Publish Ropedia Xperience-10M task baseline cards

ARTIFACT_GUIDE.md

@@ -21,13 +21,13 @@ The project separates these reading layers:
 6. **Data contract:** how one public Xperience-10M sample episode becomes
    aligned model windows and feature blocks.
 7. **Task evidence:** minimal and neural results for the 12 task contracts plus
-   audio ablation, raw-audio feature replacement, and four research-direction
+   audio contribution variants, and four research-direction
    extension probes.
 8. **Reproducibility:** public commands, expected outputs, and exact-match
    evidence for the single-episode pipeline.
 9. **Public project surface:** repo, website, and Hugging Face pages,
    accessibility semantics, links, and reader-facing copy.
-10. **Multi-episode pilot status:** scripts and reports for the planned 32-episode
+10. **Multi-episode pilot status:** scripts and reports for the selected-episode
    Qwen3-Omni pilot, with the data-access requirement kept visible.
 11. **Foundation-model selection:** Qwen3-Omni, Cosmos 3, GR00T, OpenVLA,
    openpi, Gemini Robotics, and lightweight policy candidates separated by
@@ -37,8 +37,8 @@ The project separates these reading layers:
 
 | Artifact | Why to open it first |
 | --- | --- |
-| [`PROJECT_STATUS.md`](PROJECT_STATUS.md) | Gives the fastest current-state table: implemented, data-gated, and outside current scope. |
-| [`RESEARCH_ROADMAP.md`](RESEARCH_ROADMAP.md) | Shows the staged path from public-sample task development to multi-episode data staging, the 32-episode Qwen3-Omni LoRA pilot, robustness runs, and larger omni-model extensions. |
+| [`PROJECT_STATUS.md`](PROJECT_STATUS.md) | Gives the fastest current-state table: implemented, in staging, and outside current scope. |
+| [`RESEARCH_ROADMAP.md`](RESEARCH_ROADMAP.md) | Shows the staged path from public-sample task development to multi-episode data staging, Qwen3-Omni LoRA, robustness runs, and larger omni-model extensions. |
 | [`FOUNDATION_MODEL_PLAN.md`](FOUNDATION_MODEL_PLAN.md) | Explains which foundation backbones fit which Xperience-10M objective: Qwen3-Omni first, Cosmos 3 for world modeling, and VLA/policy models after action-target conversion. |
 | [`EVIDENCE_CONTRACT.md`](EVIDENCE_CONTRACT.md) | Defines the implemented scope, setup-stage artifacts, and multi-episode prerequisites. |
 | [`QUALITY_GATES.md`](QUALITY_GATES.md) | Lists the automated release checks and post-publish verification used to keep the release current. |
@@ -64,7 +64,7 @@ The project separates these reading layers:
 | [`docs/data/live_publication_status.json`](docs/data/live_publication_status.json) | Last live GitHub/HF verification after upload. |
 | [`docs/data/mirror_parity.json`](docs/data/mirror_parity.json) | Confirms prepared HF Space, artifact, and model mirrors match the repo for critical data, figures, website HTML, and validator scripts. |
 | [`docs/data/publication_audit.json`](docs/data/publication_audit.json) | Summarizes public bundle contents and exclusions for raw data, Python caches, heavy archives, token strings, and public-card figure references. |
-| [`docs/data/scope_claims_audit.json`](docs/data/scope_claims_audit.json) | Records historical `32ep` setup identifiers separately from completed held-out-episode results. |
+| [`docs/data/scope_claims_audit.json`](docs/data/scope_claims_audit.json) | Separates setup identifiers from completed held-out-episode results. |
 | [`docs/data/task_surface_integrity.json`](docs/data/task_surface_integrity.json) | Confirms the public 12-task cards use readable task names, modality thumbnails, and the interactive walkthrough/player data contract. |
 | [`docs/data/website_integrity.json`](docs/data/website_integrity.json) | Confirms local site links, anchors, JSON bundles, and referenced images resolve. |
 | [`RENDERED_SITE_CHECK.md`](RENDERED_SITE_CHECK.md) and [`docs/data/rendered_site_check.json`](docs/data/rendered_site_check.json) | Records the latest browser-level page load, tab navigation, walkthrough deep link, player interaction, and console-health check. |
@@ -107,7 +107,7 @@ The project separates these reading layers:
 | Artifact | What it shows |
 | --- | --- |
 | [`results/episode_task_suite/windows.csv`](results/episode_task_suite/windows.csv) | The sample episode is converted into 1,161 aligned 20-frame windows. |
-| [`results/episode_task_suite/feature_manifest.json`](results/episode_task_suite/feature_manifest.json) | The current input vector has 8,546 dimensions with explicit feature-block boundaries, including a 168-d AAC audio block. |
+| [`results/episode_task_suite/feature_manifest.json`](results/episode_task_suite/feature_manifest.json) | The current input vector has 8,546 dimensions with explicit modality-group boundaries, including a 168-d audio group. |
 | [`results/episode_task_suite/available_modalities.json`](results/episode_task_suite/available_modalities.json) | The sample modality coverage is recorded, including the current audio-featurization status. |
 | [`results/audio_ablation/raw_logmel_fisheye_cam0_sr16000_mels64_fft512_hop160.npz`](results/audio_ablation/raw_logmel_fisheye_cam0_sr16000_mels64_fft512_hop160.npz) | Derived 588-d raw log-mel window features decoded from the local public-sample MP4 audio stream; raw audio itself is not redistributed. |
 | [`docs/data/modality_atlas.json`](docs/data/modality_atlas.json) | The responsive website modality cards and derived thumbnail assets are documented without redistributing raw data. |
@@ -122,9 +122,9 @@ The project separates these reading layers:
 | [`results/episode_task_suite/research_directions/`](results/episode_task_suite/research_directions/) | Mapping from the 12 tasks to the four Ropedia research directions. |
 | [`results/episode_task_suite/research_direction_extensions/`](results/episode_task_suite/research_direction_extensions/) | Four additional coded probes, one per research direction. |
 | [`results/episode_task_suite/task_walkthroughs/`](results/episode_task_suite/task_walkthroughs/) | Human-readable research names and case studies explaining input, process modules, output, metric, limitation, and the website task-player data. |
-| [`results/audio_ablation/audio_ablation_metrics.csv`](results/audio_ablation/audio_ablation_metrics.csv) | All 72 measured audio rows: 12 tasks times six variants, including no-audio, handcrafted-audio-only, raw-audio-only, raw replacement, and all-plus-raw. |
+| [`results/audio_ablation/audio_ablation_metrics.csv`](results/audio_ablation/audio_ablation_metrics.csv) | All 72 measured audio rows: 12 tasks times six variants, including no-audio, audio-only, alternate-audio-only, representation replacement, and all-input variants. |
 | [`results/audio_ablation/audio_delta_summary.csv`](results/audio_ablation/audio_delta_summary.csv) | Compact per-task audio delta table for quick manual inspection. |
-| [`scripts/audio_ablation_and_raw_upgrade.py`](scripts/audio_ablation_and_raw_upgrade.py) | Regenerates current-AAC audio ablation and raw log-mel upgrade results from real task-suite artifacts plus the local public-sample MP4. |
+| [`scripts/audio_ablation_and_raw_upgrade.py`](scripts/audio_ablation_and_raw_upgrade.py) | Regenerates audio contribution results from real task-suite artifacts plus the local public-sample MP4. |
 | [`scripts/validate_task_surface.py`](scripts/validate_task_surface.py) | Fails publication if public task cards drift back to raw artifact ids or lose their thumbnail/player wiring. |
 
 ## Reproducibility
@@ -155,8 +155,8 @@ The project separates these reading layers:
 
 | Artifact | Current status |
 | --- | --- |
-| [`results/omni_finetune/DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md) | Summarizes the data-access requirement before the 32-episode Qwen3-Omni pilot can run. |
-| [`results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`](results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md) | Documents the public multi-episode access path, selected 32-episode pilot plan, and data requirements. |
+| [`results/omni_finetune/DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md) | Summarizes the staging requirement before the held-out Qwen3-Omni pilot can report metrics. |
+| [`results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`](results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md) | Documents the public multi-episode access path, selected relay plan, and data requirements. |
 | [`scripts/omni/discover_xperience10m_sources.py`](scripts/omni/discover_xperience10m_sources.py) | Discovery gate for valid multi-episode Xperience-10M sources. |
 | [`scripts/omni/train_qwen3_omni_lora.py`](scripts/omni/train_qwen3_omni_lora.py) | Training entrypoint for the Qwen3-Omni LoRA pilot after the data gate passes. |
 | [`FOUNDATION_MODEL_PLAN.md`](FOUNDATION_MODEL_PLAN.md) | Adds the post-data-gate backbone selection plan: Qwen3-Omni first, Cosmos 3 for world modeling, and OpenVLA/openpi/GR00T for policy/action branches. |

EVIDENCE_CONTRACT.md

@@ -13,7 +13,7 @@ the dashboard as a basis for further work.
 | Public figures are indexed as project evidence. | `FIGURE_INDEX.md`, `docs/data/figure_index.json`, `scripts/build_figure_index.py` | Verified visual evidence | Derived figures and thumbnails only; does not include raw MP4/HDF5/RRD data |
 | The project logo is consistently packaged across public surfaces. | `docs/data/brand_assets.json`, `docs/assets/brand/`, `scripts/build_brand_assets.py` | Verified brand packaging | Generated presentation assets only; does not contain raw Xperience-10M data or model weights |
 | The public Xperience-10M sample has been converted into aligned model windows. | `results/episode_task_suite/windows.csv`, `results/episode_task_suite/shared_windows.npz`, `results/episode_task_suite/summary_report.json` | Verified for 5,821 frames and 1,161 windows | One public sample episode only |
-| The current feature contract is explicit and inspectable. | `results/episode_task_suite/feature_manifest.json`, `results/episode_task_suite/available_modalities.json` | Verified for an 8,546-d feature vector | AAC audio is decoded from `fisheye_cam0.mp4` into a 168-d feature block |
+| The current feature contract is explicit and inspectable. | `results/episode_task_suite/feature_manifest.json`, `results/episode_task_suite/available_modalities.json` | Verified for an 8,546-d feature vector | Synchronized video, audio, depth, pose/SLAM, motion, inertial, calibration, and language signals are represented |
 | The task evaluation protocol is explicit and generated from committed metrics. | `EVALUATION_PROTOCOL.md`, `docs/data/evaluation_protocol.json`, `scripts/build_evaluation_protocol.py` | Verified protocol | Defines windows, split, per-task metrics, leakage controls, and current limitations |
 | The public sample modalities are inspectable without raw data redistribution. | `docs/data/modality_atlas.json`, `docs/assets/modalities/`, website modality atlas | Verified derived thumbnail atlas | Thumbnails are presentation assets, not a replacement for official raw data access |
 | Public task cards stay readable for non-expert readers. | `docs/data/task_surface_integrity.json`, `scripts/validate_task_surface.py`, website task cards/player | Task-surface report | Presentation layer only; it does not add model quality or new data |
@@ -21,9 +21,9 @@ the dashboard as a basis for further work.
 | Minimal and neural heads use the same task contracts. | `scripts/neural_task_models.py`, `results/episode_task_suite/neural_mlp/`, `docs/assets/task_architectures.png` | Verified for 12 minimal heads and 12 neural MLP heads | Small heads only; not a foundation model |
 | Four Ropedia research directions are mapped honestly as direct, proxy, or diagnostic evidence. | `results/episode_task_suite/research_directions/research_direction_taxonomy.json`, `docs/data/research_directions.json` | Verified taxonomy | Some directions remain proxy-only |
 | Four extra direction probes are coded and evaluated. | `results/episode_task_suite/research_direction_extensions/research_direction_extension_results.json`, `docs/data/research_direction_extensions.json` | Verified single-episode probes | Not full human modeling, neural rendering, intent modeling, or world modeling solutions |
-| Qwen3-Omni infrastructure has passed setup checks. | `results/omni_finetune/RUN_REPORT.md`, `results/omni_finetune/dataset_manifest.json`, `results/omni_finetune/metrics_eval.json` | Setup-stage evidence | One episode, 128 train windows; full metrics require the 32-episode pilot |
-| The 32-episode LoRA pilot is waiting on gated data access. | `results/omni_finetune/DATA_ACCESS_STATUS.md`, `results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`, `results/omni_finetune/source_discovery.json` | Data access pending | Held-out metrics come after the data gate, manifest construction, training, and test evaluation |
-| Historical `32ep` path strings are tracked as setup-file provenance. | `scripts/validate_scope_claims.py`, `docs/data/scope_claims_audit.json` | Multi-episode pilot status | Old run/path identifiers stay separate from completed 32-episode results |
+| Qwen3-Omni infrastructure has passed setup checks. | `results/omni_finetune/RUN_REPORT.md`, `results/omni_finetune/dataset_manifest.json`, `results/omni_finetune/metrics_eval.json` | Setup-stage evidence | One episode, 128 train windows; full metrics require completed multi-episode staging and held-out evaluation |
+| The Qwen3-Omni LoRA pilot is in multi-episode staging. | `results/omni_finetune/DATA_ACCESS_STATUS.md`, `results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`, `results/omni_finetune/source_discovery.json` | Data staging | Full-dataset access is granted; held-out metrics come after selected relay, manifest construction, training, and test evaluation |
+| Older pilot path strings are tracked as setup-file provenance. | `scripts/validate_scope_claims.py`, `docs/data/scope_claims_audit.json` | Multi-episode pilot status | Run/path identifiers stay separate from completed held-out-episode results |
 | Prepared GitHub/Hugging Face mirrors carry matching critical files. | `scripts/validate_mirror_parity.py`, `docs/data/mirror_parity.json` | Mirror parity report | Compares prepared data files, visual assets, website HTML, and validator scripts before upload; live URLs are checked after publishing |
 | The public GitHub and Hugging Face bundles are ready to share. | `scripts/validate_publication_package.py`, `docs/data/publication_audit.json` | Public bundle contents | Covers public files, HF bundles, and public-card freshness; ignored local scratch outputs are excluded |
 | The public repo, website, and Hugging Face cards present one cohesive research project. | `PUBLIC_SURFACE_QA.md`, `scripts/build_public_surface_qa.py`, `docs/data/public_surface_qa.json` | Public project surface | Covers SEO/social metadata, accessible tab semantics, public links, project links, and reader-facing copy |
@@ -32,7 +32,7 @@ the dashboard as a basis for further work.
 | The release checks are explicit. | `QUALITY_GATES.md`, `scripts/build_quality_gates.py`, `docs/data/quality_gates.json` | Release checks | Summarizes packaging and live-mirror checks; cross-episode model quality is measured by later held-out reports |
 | The live public mirrors are verified after upload. | `scripts/verify_live_publication.py`, `docs/data/live_publication_status.json` | Live publication report | Fetches public GitHub/HF URLs; it does not validate private training state |
 | The core project artifacts are indexed and grouped for fast reading. | `ARTIFACT_GUIDE.md`, `scripts/build_artifact_index.py`, `docs/data/artifact_index.json` | Verified guide and index | Selective source-of-truth catalog, not a complete inventory of every output file |
-| The public reproduction path is documented. | `REPRODUCIBILITY.md`, `docs/data/reproducibility_matrix.json`, `notes/reproducibility_audit.md` | Verified documentation and prior exact-match check | Publicly reproduces the single-episode pipeline, not the gated 32-episode Qwen3-Omni pilot |
+| The public reproduction path is documented. | `REPRODUCIBILITY.md`, `docs/data/reproducibility_matrix.json`, `notes/reproducibility_audit.md` | Verified documentation and prior exact-match check | Publicly reproduces the single-episode pipeline; multi-episode Qwen3-Omni metrics are added only after staging and held-out evaluation |
 | The project is externally citable and machine-readable. | `CITATION.cff`, `codemeta.json`, `docs/data/project_manifest.json`, `LICENSE` | Verified metadata files | Code license does not override original Xperience-10M dataset terms |
 | A first-time reader has an explicit project path. | `docs/data/project_packet.json`, website project path section, README project path | Verified project packet | Guides inspection across data, tasks, results, and scale-up status |
 
@@ -69,8 +69,8 @@ the dashboard as a basis for further work.
    modalities enter the current feature vector.
 13. Inspect `results/episode_task_suite/neural_mlp/` to compare minimal and
    neural heads under the same splits.
-14. Inspect `docs/data/scope_claims_audit.json` before interpreting historical
-   `32ep` strings in Qwen3-Omni setup artifacts.
+14. Inspect `docs/data/scope_claims_audit.json` before interpreting older
+   Qwen3-Omni setup artifacts.
 15. Inspect `docs/data/mirror_parity.json` before assuming the GitHub and
    Hugging Face mirrors contain the same critical data, visual, HTML, and
    validator files.

FOUNDATION_MODEL_PLAN.md

@@ -13,13 +13,13 @@ run a held-out multi-episode foundation-model evaluation.
 
 | Priority | Model family | Best role for this project | Why it fits Xperience-10M | Current decision |
 | --- | --- | --- | --- | --- |
-| 1 | Qwen3-Omni | Multimodal instruction model and JSON task predictor | Accepts video/audio/language directly; depth, pose, mocap, and IMU can enter through the existing sensor bridge | Keep as first 32-episode LoRA pilot |
+| 1 | Qwen3-Omni | Multimodal instruction model and JSON task predictor | Accepts video/audio/language directly; depth, pose, mocap, and IMU can enter through the existing sensor bridge | Keep as the first selected-episode LoRA pilot |
 | 2 | Cosmos 3 | Embodied world model, action generation, and synthetic future prediction | Designed for physical-world video generation, action-conditioned world modeling, and robot/world simulation style objectives | Add as the first world-model branch after the data gate |
 | 3 | NVIDIA GR00T | Humanoid/action-policy foundation model | Xperience-10M mocap, hand motion, contacts, and egocentric interaction can support retargeting and action-understanding probes | Track as a humanoid policy branch, not the first LoRA pilot |
 | 4 | OpenVLA / OpenVLA-OFT | Open vision-language-action policy baseline | Useful when windows are converted into visual observation plus action-token targets | Use after action-space design is explicit |
 | 5 | openpi pi0/pi0.5 | Open robot policy and action expert baseline | Useful for action chunking, policy fine-tuning, and embodiment transfer experiments | Candidate for policy branch once action labels are retargeted |
 | 6 | Gemini Robotics | Closed/API embodied reasoning reference | Strong candidate for qualitative reasoning and task interpretation, but not a local fine-tune target | Use only as an external comparison or annotation assistant |
-| 7 | Octo / SmolVLA-style lightweight policies | Smaller reproducible robot-policy baselines | Good for cheaper action-policy experiments, but less directly omni-modal | Optional baseline branch after 32-episode data staging |
+| 7 | Octo / SmolVLA-style lightweight policies | Smaller reproducible robot-policy baselines | Good for cheaper action-policy experiments, but less directly omni-modal | Optional baseline branch after selected-episode data staging |
 
 ## Why Qwen3-Omni Still Goes First
 
@@ -95,7 +95,7 @@ The foundation-model stage should add metrics beyond the current 12-task suite:
 
 ## Execution Order
 
-1. Finish multi-episode data staging with at least 32 valid episodes.
+1. Finish multi-episode data staging for the selected relay.
 2. Run the Qwen3-Omni LoRA pilot exactly once as the first held-out baseline.
 3. Run a model-selection dry run on 3-8 episodes: Qwen3-Omni prompt-only,
    Qwen3-Omni LoRA, Cosmos 3 world-model preprocessing, and one policy baseline.

PROJECT_BRIEF.md

@@ -10,11 +10,11 @@ egocentric episode before scaling to held-out multi-episode training?
 | Layer | Current artifact |
 | --- | --- |
 | Data unit | 1 public sample episode, 5,821 frames, 1,161 synchronized 20-frame windows |
-| Modalities | Video-derived features, AAC audio, depth, pose/SLAM, mocap, IMU, calibration, and language-derived features |
+| Modalities | Video-derived features, audio, depth, pose/SLAM, mocap, IMU, calibration, and language-derived features |
 | Task suite | 12 embodied-AI task contracts with inputs, targets, metrics, predictions, and case-study walkthroughs |
 | Models | Minimal linear/ridge/logistic baselines plus compact PyTorch MLP heads for the same 12 tasks |
 | Research map | Four Ropedia research directions with direct, proxy, diagnostic, and extension-task coverage |
-| Scale-up path | Qwen3-Omni LoRA pilot code path prepared for 32 held-out episodes after gated data access |
+| Scale-up path | Qwen3-Omni LoRA code path prepared; full-dataset access is granted and a 128-episode selected relay is being staged |
 
 ## How To Read It
 

PROJECT_README.md

@@ -31,7 +31,7 @@ For a first pass, use [`PROJECT_BRIEF.md`](PROJECT_BRIEF.md) or the
 machine-readable [`docs/data/project_brief.json`](docs/data/project_brief.json).
 They give the project shape in one page: what exists now, what the public
 sample can support, where the 12 tasks and baselines live, and what must happen
-before the 32-episode omni-model stage becomes a real held-out evaluation.
+before the multi-episode omni-model stage becomes a real held-out evaluation.
 
 | Reader goal | Best entry point |
 | --- | --- |
@@ -39,6 +39,7 @@ before the 32-episode omni-model stage becomes a real held-out evaluation.
 | See the visual research dashboard | [GitHub Pages dashboard](https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/) |
 | Navigate the 12 tasks, four tracks, and scale-up plan | [Interactive research roadmap](https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/research_roadmap.html), [`docs/data/research_roadmap_interactive.json`](docs/data/research_roadmap_interactive.json) |
 | Compare current task metrics | [`RESEARCH_TAKEAWAYS.md`](RESEARCH_TAKEAWAYS.md), [`docs/data/summary_metrics.json`](docs/data/summary_metrics.json) |
+| Compare possible foundation backbones | [`FOUNDATION_MODEL_PLAN.md`](FOUNDATION_MODEL_PLAN.md), [`docs/data/foundation_model_plan.json`](docs/data/foundation_model_plan.json) |
 | Understand one model input | [`results/episode_task_suite/feature_manifest.json`](results/episode_task_suite/feature_manifest.json), [`results/episode_task_suite/windows.csv`](results/episode_task_suite/windows.csv) |
 | Check multi-episode data status | [`results/omni_finetune/DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md) |
 
@@ -46,12 +47,12 @@ before the 32-episode omni-model stage becomes a real held-out evaluation.
 
 | Theme | Current implementation |
 | --- | --- |
-| Dataset slice | One public Xperience-10M sample episode, 5,821 frames, 1,161 windows, and 8,546 extracted feature dimensions |
-| Modalities | Video-derived features, AAC audio features, depth, camera pose/SLAM, hand/body mocap, IMU, calibration, and language-derived features |
+| Dataset slice | One public Xperience-10M sample episode, 5,821 frames, 1,161 windows, and an 8,546-dimensional representation |
+| Modalities | Video, audio, depth, camera pose/SLAM, hand/body mocap, IMU, calibration, and language annotations |
 | Task suite | 12 human-readable embodied-AI task contracts with input, process, output, metrics, predictions, and case-study walkthroughs |
 | Baselines | Minimal linear/ridge/logistic heads plus compact PyTorch MLP task heads over the same chronological split |
 | Research directions | Task mapping and extension probes for human modeling, 3D/4D reconstruction, egocentric interaction, and world modeling |
-| Scale-up path | Data-gated Qwen3-Omni LoRA pilot plan for 32 held-out episodes; moves to completed once data, training, and held-out evaluation are present |
+| Scale-up path | Full-dataset access granted; a 128-episode selected relay is being staged with chunked parallel transfer and overlapping batch prefetch before Qwen3-Omni LoRA, followed by Cosmos 3/world-model and VLA/policy branches |
 | Public surfaces | GitHub repo, GitHub Pages dashboard, HF Space, HF artifact dataset, HF baseline-model repo, and HF collection |
 
 For the fastest interpretation of the current metrics, start with
@@ -71,8 +72,8 @@ Current contributions:
 - a generated four-direction research taxonomy matching the Ropedia job tracks,
 - four additional direction-extension probes with minimal and neural baselines,
 - human-readable research task cards and an interactive scrub/play walkthrough storyboard for every task,
-- an interactive research roadmap connecting 12 tasks, four research tracks, current sample evidence, and the Qwen3-Omni scale-up path,
-- a next-milestone track for Qwen3-Omni fine-tuning and sensor-bridge evaluation,
+- an interactive research roadmap connecting 12 tasks, four research tracks, current sample evidence, the Qwen3-Omni scale-up path, and foundation-model branch selection,
+- a next-milestone track for Qwen3-Omni fine-tuning, Cosmos 3 world modeling, and sensor-bridge evaluation,
 - metrics, predictions, model weights, manifests, charts, and a two-level
   tabbed static research website,
 - a clear explanation of what is implemented now and what moves to the multi-episode stage.
@@ -89,11 +90,12 @@ multi-episode held-out model metrics:
 | Figure index | `FIGURE_INDEX.md`, `docs/data/figure_index.json`, `scripts/build_figure_index.py` | catalogs public figures, charts, modality thumbnails, dimensions, hashes, roles, and source scripts |
 | Brand assets | `docs/assets/brand/`, `docs/favicon.png`, `docs/apple-touch-icon.png`, `scripts/build_brand_assets.py` | applies the generated project logo system across the website, README, HF cards, favicon, and social previews |
 | Data windows | `results/episode_task_suite/windows.csv`, `shared_windows.npz`, `summary_report.json` | one public sample episode |
-| Feature contract | `results/episode_task_suite/feature_manifest.json`, `available_modalities.json` | 8,546 current features, including a real AAC audio block decoded from `fisheye_cam0.mp4` |
+| Feature contract | `results/episode_task_suite/feature_manifest.json`, `available_modalities.json` | documents the 8,546-dimensional multimodal representation and source coverage |
 | Evaluation protocol | `EVALUATION_PROTOCOL.md`, `docs/data/evaluation_protocol.json`, `scripts/build_evaluation_protocol.py` | defines windowing, chronological split, leakage controls, per-task metrics, and current limitations |
 | Research takeaways | `RESEARCH_TAKEAWAYS.md`, `docs/data/research_takeaways.json`, `scripts/build_research_takeaways.py` | summarizes result interpretation from committed metrics and identifies which experiments need held-out episodes |
-| Audio ablation | `scripts/audio_ablation_and_raw_upgrade.py`, `results/audio_ablation/`, `docs/data/audio_ablation_summary.json` | measures current AAC audio contribution and a raw log-mel audio replacement across all 12 task contracts |
-| Research roadmap | `RESEARCH_ROADMAP.md`, `docs/research_roadmap.html`, `docs/data/research_roadmap.json`, `docs/data/research_roadmap_interactive.json` | stages and visualizes the path from public-sample task development to multi-episode held-out evaluation and larger omni-model extensions |
+| Audio ablation | `scripts/audio_ablation_and_raw_upgrade.py`, `results/audio_ablation/`, `docs/data/audio_ablation_summary.json` | measures whether audio helps each of the 12 task contracts |
+| Research roadmap | `RESEARCH_ROADMAP.md`, `docs/research_roadmap.html`, `docs/data/research_roadmap.json`, `docs/data/research_roadmap_interactive.json` | stages and visualizes the path from public-sample task development to multi-episode held-out evaluation, foundation-model selection, and larger omni/world-model extensions |
+| Foundation-model plan | `FOUNDATION_MODEL_PLAN.md`, `docs/data/foundation_model_plan.json` | keeps Qwen3-Omni as the first trainable pilot, adds Cosmos 3 as the first world-model branch, and tracks OpenVLA/openpi/GR00T policy candidates |
 | 12-task suite | `scripts/episode_task_suite.py`, per-task `metrics.json`, predictions | chronological single-episode split |
 | Single-episode diagnostics | `scripts/single_episode_diagnostics.py`, `results/single_episode_diagnostics/`, `docs/single_episode_explorer.html` | modality ablations, timeline overlay, object-label export, alignment stress tests, and interactive window inspection from one sample episode |
 | Neural heads | `scripts/neural_task_models.py`, `results/episode_task_suite/neural_mlp/` | compact MLP heads, not a foundation model |
@@ -101,8 +103,8 @@ multi-episode held-out model metrics:
 | Task surface integrity | `docs/data/task_surface_integrity.json`, `scripts/validate_task_surface.py` | public task cards stay human-readable, thumbnail-backed, and wired to the scrub/play walkthrough storyboard |
 | Rendered website check | `RENDERED_SITE_CHECK.md`, `docs/data/rendered_site_check.json`, `scripts/build_rendered_site_check.py` | records a browser-level load, tab, walkthrough deep-link, control-click, and console-health check |
 | Public project surface | `PUBLIC_SURFACE_QA.md`, `docs/data/public_surface_qa.json`, `scripts/build_public_surface_qa.py` | presents the repo, website, and Hugging Face cards as one research project surface |
-| Qwen3-Omni | `results/omni_finetune/DATA_ACCESS_STATUS.md`, `MULTI_EPISODE_ACCESS_STATUS.md` | setup-stage until 32 valid episodes are available and held-out evaluation runs |
-| Multi-episode pilot status | `scripts/validate_scope_claims.py`, `docs/data/scope_claims_audit.json` | records setup-stage `32ep` artifacts separately from completed held-out-episode metrics |
+| Qwen3-Omni | `results/omni_finetune/DATA_ACCESS_STATUS.md`, `MULTI_EPISODE_ACCESS_STATUS.md` | full-dataset access is granted; 128 selected episodes are in accelerated relay/staging before held-out evaluation |
+| Multi-episode pilot status | `scripts/validate_scope_claims.py`, `docs/data/scope_claims_audit.json` | separates setup artifacts, selected relay state, and completed held-out-episode metrics |
 | Mirror parity | `scripts/validate_mirror_parity.py`, `docs/data/mirror_parity.json` | prepared GitHub/HF mirrors carry matching data, figure, website HTML, and validator files |
 | Public bundle contents | `scripts/validate_publication_package.py`, `docs/data/publication_audit.json` | summarizes the public repo and HF bundles, including raw-data exclusion and local scratch-file exclusion |
 | Release checks | `QUALITY_GATES.md`, `docs/data/quality_gates.json`, `scripts/build_quality_gates.py` | one map for automated checks and live post-publish verification |
@@ -139,6 +141,9 @@ The generated research takeaways are at
 The staged research roadmap is at
 [`RESEARCH_ROADMAP.md`](RESEARCH_ROADMAP.md) and
 [`docs/data/research_roadmap.json`](docs/data/research_roadmap.json).
+The foundation-model selection plan is at
+[`FOUNDATION_MODEL_PLAN.md`](FOUNDATION_MODEL_PLAN.md) and
+[`docs/data/foundation_model_plan.json`](docs/data/foundation_model_plan.json).
 The source-of-truth artifact index is at
 [`docs/data/artifact_index.json`](docs/data/artifact_index.json).
 For a human-readable artifact map, use
@@ -173,14 +178,14 @@ They give the current research state in one compact table:
 
 | Area | Current decision |
 | --- | --- |
-| Public-sample pipeline | Verified on one public sample episode: 5,821 frames, 1,161 windows, 8,546 current features |
+| Public-sample pipeline | Verified on one public sample episode: 5,821 frames, 1,161 windows, 8,546 dimensions |
 | 12-task suite | Verified minimal baselines with committed metrics, predictions, and manifests |
 | Neural heads | Verified compact PyTorch MLP heads over the same task contracts and chronological splits |
 | Official dataset wording | Verified against the public `ropedia-ai/xperience-10m` dataset card/API metadata |
 | Source alignment | Source facts, sample details, API-listing notes, and project coverage are consistent across repo, website, and HF cards |
 | Evaluation protocol | Verified generated protocol for windowing, split policy, leakage controls, and per-task metrics |
 | Website and HF mirrors | Verified by website reference reports, public project-surface reports, mirror parity, and live-publication checks; the public dashboard uses five top-level tabs plus subsection tabs for dataset, task-suite, method, result, and resource views |
-| Qwen3-Omni 32-episode pilot | Data-gated; prepared, with full metrics pending held-out evaluation |
+| Qwen3-Omni multi-episode pilot | Full-dataset access granted; 128-episode relay in progress with chunked parallel transfer and batch prefetch, with full metrics pending completed staging and held-out evaluation |
 | Raw Xperience-10M data / full Qwen weights | Not redistributed |
 
 ## 90-Second Research Project Path
@@ -194,12 +199,13 @@ If you are reading the project cold, open these in order:
 | 3 | Are source facts consistently presented? | [`SOURCE_ALIGNMENT_AUDIT.md`](SOURCE_ALIGNMENT_AUDIT.md), [`docs/data/source_alignment_audit.json`](docs/data/source_alignment_audit.json), [`scripts/validate_source_alignment.py`](scripts/validate_source_alignment.py) | Repo, website, and HF cards use the same full-dataset facts, sample-card facts, API-listing notes, and project coverage. |
 | 4 | How exactly are tasks evaluated? | [`EVALUATION_PROTOCOL.md`](EVALUATION_PROTOCOL.md), [`docs/data/evaluation_protocol.json`](docs/data/evaluation_protocol.json), [`scripts/build_evaluation_protocol.py`](scripts/build_evaluation_protocol.py) | The window unit, chronological split, leakage controls, task metrics, and current limitations are explicit. |
 | 5 | What do the current results mean? | [`RESEARCH_TAKEAWAYS.md`](RESEARCH_TAKEAWAYS.md), [`docs/data/research_takeaways.json`](docs/data/research_takeaways.json), [`docs/data/summary_metrics.json`](docs/data/summary_metrics.json) | The takeaways are generated from committed metrics and identify which signals are ready for larger held-out experiments. |
-| 6 | What is the staged roadmap? | [`RESEARCH_ROADMAP.md`](RESEARCH_ROADMAP.md), [`docs/data/research_roadmap.json`](docs/data/research_roadmap.json), [`DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md) | The roadmap connects public-sample task development to multi-episode staging, Qwen3-Omni LoRA, robustness runs, and larger omni-model extensions. |
-| 7 | How do I reproduce it? | [`REPRODUCIBILITY.md`](REPRODUCIBILITY.md), [`docs/data/reproducibility_matrix.json`](docs/data/reproducibility_matrix.json), [`notes/reproducibility_audit.md`](notes/reproducibility_audit.md) | Public commands, expected outputs, and the latest exact-match reproduction record are explicit. |
-| 8 | What is one model input? | [`windows.csv`](results/episode_task_suite/windows.csv), [`feature_manifest.json`](results/episode_task_suite/feature_manifest.json), [`available_modalities.json`](results/episode_task_suite/available_modalities.json) | The input is an aligned 8,546-d window vector with explicit feature-block boundaries. |
-| 9 | Are the task results backed by files? | [`summary_report.json`](results/episode_task_suite/summary_report.json), [`neural_mlp/`](results/episode_task_suite/neural_mlp/), [`docs/data/summary_metrics.json`](docs/data/summary_metrics.json) | Each task has minimal and neural-head evidence over the same window contracts. |
-| 10 | Is the website self-consistent? | [`docs/data/website_integrity.json`](docs/data/website_integrity.json), [`scripts/validate_website_integrity.py`](scripts/validate_website_integrity.py) | Local links, anchors, tab routing, JSON data, and referenced images are checked before publishing. |
-| 11 | What is still pending? | [`DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md), [`MULTI_EPISODE_ACCESS_STATUS.md`](results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md), [`scripts/omni/discover_xperience10m_sources.py`](scripts/omni/discover_xperience10m_sources.py) | The 32-episode Qwen3-Omni run is prepared; final model metrics require gated data and held-out evaluation. |
+| 6 | What is the staged roadmap? | [`RESEARCH_ROADMAP.md`](RESEARCH_ROADMAP.md), [`docs/data/research_roadmap.json`](docs/data/research_roadmap.json), [`DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md) | The roadmap connects public-sample task development to multi-episode staging, Qwen3-Omni LoRA, foundation-model selection, robustness runs, and larger omni/world-model extensions. |
+| 7 | Which foundation model comes next? | [`FOUNDATION_MODEL_PLAN.md`](FOUNDATION_MODEL_PLAN.md), [`docs/data/foundation_model_plan.json`](docs/data/foundation_model_plan.json) | Qwen3-Omni remains the first held-out LoRA baseline; Cosmos 3 is the first world-model branch; OpenVLA/openpi/GR00T wait for explicit action targets. |
+| 8 | How do I reproduce it? | [`REPRODUCIBILITY.md`](REPRODUCIBILITY.md), [`docs/data/reproducibility_matrix.json`](docs/data/reproducibility_matrix.json), [`notes/reproducibility_audit.md`](notes/reproducibility_audit.md) | Public commands, expected outputs, and the latest exact-match reproduction record are explicit. |
+| 9 | What is one model input? | [`windows.csv`](results/episode_task_suite/windows.csv), [`feature_manifest.json`](results/episode_task_suite/feature_manifest.json), [`available_modalities.json`](results/episode_task_suite/available_modalities.json) | The input is an aligned 8,546-dimensional multimodal window with synchronized video, audio, sensor, and language signals. |
+| 10 | Are the task results backed by files? | [`summary_report.json`](results/episode_task_suite/summary_report.json), [`neural_mlp/`](results/episode_task_suite/neural_mlp/), [`docs/data/summary_metrics.json`](docs/data/summary_metrics.json) | Each task has minimal and neural-head evidence over the same window contracts. |
+| 11 | Is the website self-consistent? | [`docs/data/website_integrity.json`](docs/data/website_integrity.json), [`scripts/validate_website_integrity.py`](scripts/validate_website_integrity.py) | Local links, anchors, tab routing, JSON data, and referenced images are checked before publishing. |
+| 12 | What is still pending? | [`DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md), [`MULTI_EPISODE_ACCESS_STATUS.md`](results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md), [`scripts/omni/discover_xperience10m_sources.py`](scripts/omni/discover_xperience10m_sources.py) | The multi-episode Qwen3-Omni run is prepared at the selection and relay level; final model metrics require completed staging, preprocessing, training, and held-out evaluation. |
 
 The machine-readable project packet is
 [`docs/data/project_packet.json`](docs/data/project_packet.json).
@@ -225,11 +231,11 @@ generated from committed metric artifacts. They define:
 - the 20-frame window unit, stride, feature dimension, and raw-data policy,
 - the chronological 70/30 single-episode split and its generalization limit,
 - the per-task input, target, primary metric, minimal score, and neural score,
-- leakage controls for future labels, target feature blocks, caption/object
+- leakage controls for future labels, target-side signals, caption/object
   labels, and train-only normalization,
 - current limitations, including cross-episode generalization,
-  audio-visual learning, pixel-depth reconstruction, and real 32-episode
-  Qwen3-Omni quality.
+  audio-visual learning, pixel-depth reconstruction, and real held-out
+  multi-episode Qwen3-Omni quality.
 
 ## Official Dataset Alignment
 
@@ -262,22 +268,22 @@ The public sample repo,
 is separately documented as `Xperience-10M-Sample` with sample metadata,
 `cc-by-nc-4.0` license, HOMIE Toolkit usage, and Rerun 0.29.0 `.rrd`
 visualization. This project preserves that distinction: the sample powers the
-current 5,821-frame task suite, while the full gated dataset remains the
-future source for held-out multi-episode training.
+current 5,821-frame task suite, while the full gated dataset is the source for
+the selected 128-episode held-out multi-episode relay now in progress.
 
 This repo's current verified subset is much smaller and intentionally explicit:
 
 - one public sample episode, 5,821 frames, and 1,161 aligned windows,
-- raw sample files with six MP4 video streams and AAC audio streams,
+- raw sample files with six MP4 video streams and audio streams,
 - `annotation.hdf5` carrying depth, SLAM/camera pose, hand/body mocap, IMU,
   language/caption annotations, calibration, metadata, and timing records,
-- an 8,546-d baseline feature vector using video-derived statistics, AAC audio,
-  depth, pose/SLAM, mocap, IMU, calibration, and language-derived blocks.
+- an 8,546-dimensional baseline representation using video, audio, depth,
+  pose/SLAM, mocap, IMU, calibration, and language-derived signals.
 
 The same alignment note also records what is outside the current implemented subset: real
 audio-visual learning, caption generation, pixel-depth estimation, SLAM
 estimation, neural rendering, policy learning, cross-episode generalization,
-and real 32-episode Qwen3-Omni model quality.
+and real held-out multi-episode Qwen3-Omni model quality.
 It also preserves the official responsible-use scope: the open-source
 dataset is limited in diversity and showcase/production quality, and it should
 not be used for identity recognition, re-identification, biometric profiling,
@@ -542,7 +548,7 @@ python scripts/train_all_modalities_model.py --workspace /path/to/workspace
 
 This repo includes a first Qwen3-Omni fine-tuning path over Xperience-10M. The
 current artifacts are setup-stage evidence, with held-out multi-episode metrics
-pending gated data access.
+pending completed staging, preprocessing, training, and evaluation.
 The useful distinction is:
 
 - direct Qwen3-Omni inputs: RGB/fisheye video, embedded MP4 audio, and language
@@ -552,9 +558,9 @@ The useful distinction is:
 
 The current scale-up artifacts show that the export, manifest, sensor-feature,
 LoRA, and evaluation scripts can run on the available sample episode. They do
-do not show a real 32-episode result. A real pilot requires at least 32 valid
+not show a real multi-episode result. A real pilot requires staged valid
 episodes, held-out episode splits, training metadata, predictions, metrics, and
-a run report.
+a run report; the current selected relay target is 128 episodes.
 
 ### Sample Count Decision
 
@@ -579,7 +585,7 @@ python scripts/omni/plan_finetune_sample_budget.py \
   --full-preview-per-episode-gb 5.1
 ```
 
-### 32-Episode Readiness Gate
+### Multi-Episode Readiness Gate
 
 ```bash
 python scripts/omni/discover_xperience10m_sources.py \
@@ -590,20 +596,21 @@ python scripts/omni/discover_xperience10m_sources.py \
 
 Current status in this repo:
 
-- local_valid_episodes: 1 (degraded-valid: annotation + fisheye_cam0.mp4)
-- local_complete_episodes: 0
-- ready_for_32_episode_pilot: false
-- planned 32-episode pilot: stratified across 32 top-level session UUIDs
-- full-dataset access: gated Xperience-10M approval is still pending
+- public_sample_valid_episodes: 1 (degraded-valid: annotation + fisheye_cam0.mp4)
+- gated_metadata_audit: 12,102 complete visible episodes across 802 complete sessions
+- selected_relay_plan: 128 metadata-balanced episodes, 96/16/16 train/val/test
+- selected_download_size: 277.71 GiB excluding `visualization.rrd`
+- ready_for_held_out_pilot: false until the selected episodes are fully staged and checked
+- full-dataset access: granted; raw multi-episode staging is in progress with chunked parallel transfer and overlapping batch prefetch
 - source_discovery: `results/omni_finetune/source_discovery.json`
 - data_status: `results/omni_finetune/DATA_ACCESS_STATUS.md`
 - access_status: `results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`
 
-Use this gate before scheduling any 32-episode full fine-tune run. The pilot
-should use stratified selection, not the first 32 paths in repository order.
-The current selection plan scans 64 top-level session UUIDs, filters for
-complete leaf episodes, excludes `visualization.rrd`, applies a `0.25 GB`
-minimum episode size, and selects 32 episodes from 32 different session UUIDs.
+Use this gate before scheduling any full fine-tune run. The pilot should use
+balanced held-out selection, not the first paths in repository order. The
+current 128-episode selection filters for complete leaf episodes, excludes
+`visualization.rrd`, balances episode-size bands, and preserves one selected
+episode per top-level session UUID.
 
 ### Uploading the pilot Qwen3-Omni LoRA
 
@@ -619,6 +626,24 @@ python3 scripts/omni/upload_qwen3_omni_lora_to_hf.py \
 This script requires a valid Hugging Face token via `HF_TOKEN` or `--token`.
 Network availability to `huggingface.co` is required.
 
+### Foundation Backbone Plan
+
+The next modeling plan tracks several foundation-model branches instead of
+assuming one backbone solves every Xperience-10M objective.
+
+| Branch | Current role | When to use it |
+| --- | --- | --- |
+| Qwen3-Omni | First trainable multimodal LoRA pilot | Use for the selected 128-episode held-out baseline over video/audio/language plus sensor-bridge features. |
+| Cosmos 3 | First world-model/action-generation branch | Use after data staging for future-window prediction, action-conditioned world modeling, and synthetic-data usefulness tests. |
+| GR00T | Humanoid/action-policy branch | Use after mocap/contact retargeting creates well-defined humanoid action targets. |
+| OpenVLA / openpi | Open VLA/policy baselines | Use after the project defines robot-compatible or action-token targets. |
+| Gemini Robotics | External reasoning reference | Use only for qualitative comparison or annotation support unless local trainable access exists. |
+
+See [`FOUNDATION_MODEL_PLAN.md`](FOUNDATION_MODEL_PLAN.md) and
+[`docs/data/foundation_model_plan.json`](docs/data/foundation_model_plan.json)
+for the full selection matrix, source links, and model-specific evaluation
+additions.
+
 ## Four Research Directions
 
 The 12 tasks are now organized against the four Ropedia research directions in
@@ -633,7 +658,7 @@ The taxonomy uses two current baselines for every task:
 
 | Baseline | Role |
 | --- | --- |
-| Minimal interpretable heads | Softmax, logistic, ridge, and retrieval heads over the 8,546-d window feature vector. These expose the input/output contract cleanly. |
+| Minimal interpretable heads | Softmax, logistic, ridge, and retrieval heads over the 8,546-dimensional multimodal representation. These expose the input/output contract cleanly. |
 | Neural MLP heads | Small PyTorch MLP classifiers/regressors on the same features and splits. These check whether nonlinear heads help before moving to Qwen/Omni fine-tuning. |
 
 Current direction-level coverage:
@@ -726,7 +751,7 @@ models.
 Shared setup:
 
 ```text
-raw episode -> 20-frame windows, stride 5 -> 8,546-d current feature vector
+raw episode -> 20-frame windows, stride 5 -> 8,546-dimensional multimodal representation
 chronological split: first 70% train, last 30% test
 scalers are fit on train windows only
 ```
@@ -740,7 +765,7 @@ There are four reusable head families:
 | Ridge + cosine ranking | Language Grounding, Cross-Modal Retrieval | project one modality into another feature space, then rank candidates by cosine |
 | Multi-label logistic regression | Object Relevance Prediction | z-score non-caption features, sigmoid object heads, threshold at 0.5 |
 
-The optional neural run keeps the same feature vectors, leakage filters,
+The optional neural run keeps the same window representation, leakage filters,
 chronological splits, and metrics, but replaces the task heads with small
 PyTorch MLP classifiers or regressors. Its outputs live under
 [`results/episode_task_suite/neural_mlp/`](results/episode_task_suite/neural_mlp/),
@@ -756,11 +781,11 @@ The task-specific heads are:
 | Action Boundary Detection | all featurized modalities | linear softmax | steady vs action boundary |
 | Next-Action Prediction | all featurized modalities at `t` | linear softmax | action at `t+20` frames |
 | Hand Trajectory Forecasting | all featurized modalities at `t` | ridge regression | future 10-frame left/right hand joints |
-| Contact State Prediction | non-contact and non-caption feature blocks | linear softmax | any body contact |
-| Object Relevance Prediction | non-caption feature blocks | multi-label logistic | relevant object set |
+| Contact State Prediction | non-contact and non-caption signals | linear softmax | any body contact |
+| Object Relevance Prediction | non-caption signals | multi-label logistic | relevant object set |
 | Language Grounding | sensor windows projected to text space | ridge projection + cosine ranking | matching time window for text query |
 | Cross-Modal Retrieval | motion/IMU/camera projected to visual space | ridge projection + cosine ranking | matching depth/video window |
-| Cross-Modal Reconstruction | motion/IMU/camera | ridge regression | depth/video feature vector |
+| Cross-Modal Reconstruction | motion/IMU/camera | ridge regression | compressed depth/video target |
 | Temporal Order Verification | `[x_t, x_t+1, x_t+1-x_t]` | binary linear softmax | correct vs reversed order |
 | Multimodal Synchronization Detection | motion plus visual pair | binary linear softmax | aligned vs shifted by 8 windows |
 
@@ -769,7 +794,7 @@ The task-specific heads are:
 | Experiment | Main score | Accuracy | Notes |
 | --- | ---: | ---: | --- |
 | Motion-only action | 0.9688 macro-F1 | 0.9828 | Uses motion/IMU features only |
-| Current all-feature action | 0.9829 macro-F1 | 0.9863 | 8,546-dimensional feature vector |
+| Current all-feature action | 0.9829 macro-F1 | 0.9863 | 8,546-dimensional multimodal representation |
 | Motion-only subtask | 0.9528 macro-F1 | 0.9759 | Strong within-episode subtask signal |
 | Current all-feature subtask | 0.9173 macro-F1 | 0.9828 | High accuracy, lower class-balanced score |
 | Cross-modal retrieval | 0.3678 top-5 | n/a | Motion/IMU/camera/audio retrieves matching depth/video |
@@ -778,27 +803,26 @@ The task-specific heads are:
 | Neural MLP hand forecast | 0.1079 MPJPE | n/a | Same features/split, nonlinear regression head |
 | Neural MLP temporal order | 0.8520 F1 | 0.8578 | Strong improvement on adjacent-window ordering |
 | Neural MLP misalignment | 0.7153 F1 | 0.7009 | Detects shifted motion/visual/audio pairs better than the linear head |
-| Audio ablation | +0.0418 mean delta | n/a | Current AAC audio improves the primary metric on 6 of 12 task contracts |
-| Raw log-mel audio replacement | +0.0936 mean delta | n/a | Raw log-mel replacement beats current handcrafted audio on 6 of 12 task contracts |
+| Audio ablation | +0.0418 mean delta | n/a | Current audio variant improves the primary metric on 6 of 12 task contracts |
+| Alternate audio representation | +0.0936 mean delta | n/a | Alternate audio-window representation improves over the baseline audio variant on 6 of 12 task contracts |
 
-## Audio Ablation and Raw-Audio Upgrade
+## Audio Contribution Study
 
-The current AAC audio block is now tested rather than only included. The script
+The audio ablation keeps the same windows and task labels, then compares input
+variants under the same chronological split. The script
 [`scripts/audio_ablation_and_raw_upgrade.py`](scripts/audio_ablation_and_raw_upgrade.py)
-reuses the real task-suite windows, decodes the local public-sample
-`fisheye_cam0.mp4` audio stream, builds a 588-d raw log-mel window feature, and
-evaluates six variants for every task: current features, no audio,
-handcrafted-audio-only, raw-audio-only, handcrafted audio replaced by raw
-log-mel, and current features plus raw log-mel.
+reuses the real task-suite windows and evaluates six variants for
+every task: current inputs, no audio, audio-only, alternate audio-only, audio
+representation replacement, and all inputs plus the alternate audio representation.
 
 The measured single-episode result is task-specific:
 
 | Readout | Value |
 | --- | ---: |
-| Tasks where current AAC audio improves the primary metric | 6 / 12 |
+| Tasks where current audio improves the primary metric | 6 / 12 |
 | Mean current-audio delta | +0.0418 |
-| Tasks where raw log-mel replacement improves over handcrafted AAC | 6 / 12 |
-| Mean raw-replacement delta vs current audio | +0.0936 |
+| Tasks where alternate audio representation improves over baseline audio | 6 / 12 |
+| Mean alternate-representation delta vs baseline audio | +0.0936 |
 
 Full files:
 
@@ -813,7 +837,7 @@ Full files:
 The neural baseline was run locally with `--include-neural` for all 12 tasks
 using 80 epochs, hidden size 128, batch size 128, and CPU execution. It is not a
 foundation model result; it is a controlled nonlinear-head comparison over the
-same 8,546-d handcrafted window features.
+same 8,546-dimensional multimodal representation.
 
 | Task | Neural metric | Minimal metric | Readout |
 | --- | ---: | ---: | --- |
@@ -848,10 +872,10 @@ artifact-driven diagnostics pass over the public sample episode:
 - `results/single_episode_diagnostics/alignment_stress/alignment_shift_metrics.csv`
   evaluates cross-modal retrieval under explicit time shifts.
 - `docs/single_episode_explorer.html` is a static interactive page for
-  inspecting window labels, objects, predictions, feature-block statistics, and
+  inspecting window labels, objects, predictions, modality statistics, and
   diagnostic scores.
 
-These are single-episode research diagnostics. They are useful for auditing
+These are single-episode research diagnostics. They are useful for studying
 task definitions, feature behavior, and model errors before scaling to more
 episodes; they are not reported as multi-episode benchmark results.
 
@@ -878,21 +902,21 @@ The test segment contains some action/subtask labels never seen during training.
 Timeline and next-action classifiers therefore expose the core limitation of
 single-episode learning instead of hiding it behind random splits.
 
-## Feature Blocks Used
+## Modalities Used
 
-The current feature vector has 8,546 dimensions and includes:
+The current public-sample pipeline uses:
 
 - hand/body mocap joints and contact labels,
 - camera translation and rotation,
 - IMU acceleration and gyroscope traces,
 - depth confidence features,
 - six video streams,
-- AAC audio features from `fisheye_cam0.mp4`,
+- audio from the sample MP4 stream,
 - caption/object/interaction text features,
 - SLAM point-cloud summary features,
 - calibration parameters.
 
-The exact feature block boundaries are stored in
+The full technical source manifest is stored in
 [`results/episode_task_suite/feature_manifest.json`](results/episode_task_suite/feature_manifest.json).
 
 ## Data Notice

PROJECT_STATUS.md

@@ -2,17 +2,17 @@
 
 This is the fastest way to understand the current research project state.
 It summarizes what has already been implemented from the public
-Xperience-10M sample, what remains data-gated, and which artifacts support
-the next development step.
+Xperience-10M sample, what is being staged for multi-episode training, and
+which artifacts support the next development step.
 
 | Area | Current state | Evidence | Research readout |
 | --- | --- | --- | --- |
 | Public-sample pipeline | Verified | `results/episode_task_suite/summary_report.json`, `results/episode_task_suite/windows.csv`, `results/episode_task_suite/feature_manifest.json` | One public Xperience-10M sample episode is converted into 5,821 frames, 1,161 aligned 20-frame windows, and an 8,546-dimensional current feature contract. |
 | Task suite | Verified | `scripts/episode_task_suite.py`, `results/episode_task_suite/`, `docs/data/summary_metrics.json` | All 12 task contracts have committed metrics, predictions, and minimal baseline outputs. |
 | Neural heads | Verified | `scripts/neural_task_models.py`, `results/episode_task_suite/neural_mlp/` | Each task also has a compact PyTorch MLP run over the same feature tensor and chronological split. |
-| Audio ablation and raw-audio upgrade | Verified | `scripts/audio_ablation_and_raw_upgrade.py`, `results/audio_ablation/`, `docs/data/audio_ablation_summary.json` | Current AAC audio improves the primary metric on 6 of 12 task contracts; replacing the current handcrafted block with a 588-d raw log-mel feature improves over current audio on 6 of 12 tasks. |
+| Audio contribution study | Verified | `scripts/audio_ablation_and_raw_upgrade.py`, `results/audio_ablation/`, `docs/data/audio_ablation_summary.json` | Audio variants are compared across all 12 task contracts; audio improves the primary metric on 6 of 12 tasks, and a 588-d audio-window representation improves over the baseline audio variant on 6 of 12 tasks. |
 | Research takeaways | Verified | `RESEARCH_TAKEAWAYS.md`, `docs/data/research_takeaways.json`, `scripts/build_research_takeaways.py` | The main result interpretation is generated from committed metrics: chronological class shift, neural gains on dynamics/order/alignment, open retrieval/reconstruction problems, and the need for held-out episodes. |
-| Research roadmap | Current | `RESEARCH_ROADMAP.md`, `docs/data/research_roadmap.json` | The staged path connects public-sample task development to multi-episode data staging, the 32-episode Qwen3-Omni LoRA pilot, foundation-model selection, robustness runs, and larger omni/world-model extensions. |
+| Research roadmap | Current | `RESEARCH_ROADMAP.md`, `docs/data/research_roadmap.json` | The staged path connects public-sample task development to 128-episode data staging, Qwen3-Omni LoRA, foundation-model selection, robustness runs, and larger omni/world-model extensions. |
 | Foundation-model plan | Current | `FOUNDATION_MODEL_PLAN.md`, `docs/data/foundation_model_plan.json` | Qwen3-Omni remains the first trainable held-out LoRA baseline; Cosmos 3 is added as the first world-model/action-generation branch; OpenVLA/openpi/GR00T are policy candidates after action targets are explicit. |
 | Evaluation protocol | Verified | `EVALUATION_PROTOCOL.md`, `docs/data/evaluation_protocol.json`, `scripts/build_evaluation_protocol.py` | Windowing, chronological split, per-task metrics, leakage controls, and current limitations are generated from committed metric artifacts. |
 | Official dataset wording | Verified | `XPERIENCE10M_DATASET_CARD_ALIGNMENT.md`, `docs/data/xperience10m_dataset_card_alignment.json` | Public wording is aligned to the official gated Xperience-10M dataset card, public sample card, and HF API metadata, including modalities, scale, access path, sample license/tooling, and current project coverage. |
@@ -20,7 +20,7 @@ the next development step.
 | Website and HF mirrors | Verified | `docs/data/website_integrity.json`, `docs/data/rendered_site_check.json`, `docs/data/mirror_parity.json`, `docs/data/live_publication_status.json` | Local website links/assets pass, the rendered walkthrough flow has a browser-level check, prepared mirrors match, and public GitHub/HF URLs have been verified after upload. |
 | Public bundle contents | Verified | `docs/data/publication_audit.json`, `QUALITY_GATES.md`, `docs/data/quality_gates.json` | Public bundles exclude raw data, caches, heavy archives, token strings, and stale public-card copy. |
 | Reproducibility | Verified for the public sample | `REPRODUCIBILITY.md`, `docs/data/reproducibility_matrix.json`, `notes/reproducibility_audit.md` | The public sample workflow has explicit commands, expected outputs, and exact-match reproduction evidence. |
-| Qwen3-Omni fine-tuning | Data-gated; full metrics pending | `results/omni_finetune/DATA_ACCESS_STATUS.md`, `results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md` | The 32-episode LoRA pilot is prepared; final held-out metrics require gated data access, manifest construction, training, and evaluation. |
+| Qwen3-Omni fine-tuning | Data staging; full metrics pending | `results/omni_finetune/DATA_ACCESS_STATUS.md`, `results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md` | Full-dataset access is granted and a 128-episode selected relay is in progress with chunked parallel transfer and overlapping batch prefetch; final held-out metrics require completed staging, manifest construction, training, and evaluation. |
 | Raw Xperience-10M redistribution | Not included | `DATA_NOTICE.md`, `docs/data/publication_audit.json` | Raw MP4, HDF5, RRD files, private gated data, and full Qwen weights are intentionally excluded. |
 
 ## Fast Research Route
@@ -50,14 +50,12 @@ the next development step.
 
 - Cross-episode generalization is a later multi-episode evaluation target; the
   current results use one public sample episode.
-- Historical `32ep` path names refer to setup files, not completed 32-episode
+- Older pilot path names refer to setup files, not completed held-out
   training results.
 - The current reconstruction task reconstructs feature vectors, not pixel
   depth, meshes, NeRF outputs, or Gaussian splats.
-- AAC audio is decoded from `fisheye_cam0.mp4` and included in the current
-  8,546-dimensional baseline feature vector.
-- Audio is now evaluated directly: the current AAC block and a raw log-mel
-  replacement are compared across all 12 task contracts in
+- Audio is part of the current 8,546-dimensional baseline feature vector.
+- Audio contribution is evaluated across all 12 task contracts in
   `results/audio_ablation/`.
 - Foundation-model selection is now explicit: Qwen3-Omni is the immediate
   trainable pilot, Cosmos 3 is the first world-model branch, and policy models

QUALITY_GATES.md

@@ -12,7 +12,7 @@ These checks cover public packaging, project status wording, mirror parity, and
 
 | Check | Command | Report | Current status | Needs attention when |
 | --- | --- | --- | --- | --- |
-| Multi-episode pilot status | `python scripts/validate_scope_claims.py` | `docs/data/scope_claims_audit.json` | `pass` | Historical 32ep setup/provenance strings are presented as completed 32-episode metrics. |
+| Multi-episode pilot status | `python scripts/validate_scope_claims.py` | `docs/data/scope_claims_audit.json` | `pass` | Setup/provenance strings are presented as completed held-out metrics. |
 | Source alignment | `python scripts/validate_source_alignment.py` | `docs/data/source_alignment_audit.json` | `pass` | Official full-dataset facts, sample-card facts, API-listing notes, or project coverage are missing or inconsistent. |
 | Website integrity | `python scripts/validate_website_integrity.py` | `docs/data/website_integrity.json` | `pass` | Local links, anchors, JSON bundles, or referenced image assets are missing or invalid. |
 | Rendered website check | `python scripts/build_rendered_site_check.py --input /tmp/xperience_rendered_site_observations.json` | `docs/data/rendered_site_check.json` | `pass` | The local rendered site cannot load, switch tabs, deep-link to the walkthrough, update player controls, or stay console-clean. |

README.md

@@ -47,12 +47,12 @@ before the multi-episode omni-model stage becomes a real held-out evaluation.
 
 | Theme | Current implementation |
 | --- | --- |
-| Dataset slice | One public Xperience-10M sample episode, 5,821 frames, 1,161 windows, and 8,546 extracted feature dimensions |
-| Modalities | Video-derived features, AAC audio features, depth, camera pose/SLAM, hand/body mocap, IMU, calibration, and language-derived features |
+| Dataset slice | One public Xperience-10M sample episode, 5,821 frames, 1,161 windows, and an 8,546-dimensional representation |
+| Modalities | Video, audio, depth, camera pose/SLAM, hand/body mocap, IMU, calibration, and language annotations |
 | Task suite | 12 human-readable embodied-AI task contracts with input, process, output, metrics, predictions, and case-study walkthroughs |
 | Baselines | Minimal linear/ridge/logistic heads plus compact PyTorch MLP task heads over the same chronological split |
 | Research directions | Task mapping and extension probes for human modeling, 3D/4D reconstruction, egocentric interaction, and world modeling |
-| Scale-up path | Data-gated Qwen3-Omni LoRA pilot plan for 32 held-out episodes, followed by a foundation-model selection branch that adds Cosmos 3/world-model and VLA/policy candidates |
+| Scale-up path | Full-dataset access granted; a 128-episode selected relay is being staged with chunked parallel transfer and overlapping batch prefetch before Qwen3-Omni LoRA, followed by Cosmos 3/world-model and VLA/policy branches |
 | Public surfaces | GitHub repo, GitHub Pages dashboard, HF Space, HF artifact dataset, HF baseline-model repo, and HF collection |
 
 For the fastest interpretation of the current metrics, start with
@@ -78,9 +78,6 @@ Current contributions:
   tabbed static research website,
 - a clear explanation of what is implemented now and what moves to the multi-episode stage.
 
-Model-card readers can start from the **Research Takeaways** in
-`RESEARCH_TAKEAWAYS.md` and `metrics/research_takeaways.json`.
-
 ## Current Research Scope
 
 This repo separates implemented single-episode research artifacts from future
@@ -93,10 +90,10 @@ multi-episode held-out model metrics:
 | Figure index | `FIGURE_INDEX.md`, `docs/data/figure_index.json`, `scripts/build_figure_index.py` | catalogs public figures, charts, modality thumbnails, dimensions, hashes, roles, and source scripts |
 | Brand assets | `docs/assets/brand/`, `docs/favicon.png`, `docs/apple-touch-icon.png`, `scripts/build_brand_assets.py` | applies the generated project logo system across the website, README, HF cards, favicon, and social previews |
 | Data windows | `results/episode_task_suite/windows.csv`, `shared_windows.npz`, `summary_report.json` | one public sample episode |
-| Feature contract | `results/episode_task_suite/feature_manifest.json`, `available_modalities.json` | 8,546 current features, including a real AAC audio block decoded from `fisheye_cam0.mp4` |
+| Feature contract | `results/episode_task_suite/feature_manifest.json`, `available_modalities.json` | documents the 8,546-dimensional multimodal representation and source coverage |
 | Evaluation protocol | `EVALUATION_PROTOCOL.md`, `docs/data/evaluation_protocol.json`, `scripts/build_evaluation_protocol.py` | defines windowing, chronological split, leakage controls, per-task metrics, and current limitations |
 | Research takeaways | `RESEARCH_TAKEAWAYS.md`, `docs/data/research_takeaways.json`, `scripts/build_research_takeaways.py` | summarizes result interpretation from committed metrics and identifies which experiments need held-out episodes |
-| Audio ablation | `scripts/audio_ablation_and_raw_upgrade.py`, `results/audio_ablation/`, `docs/data/audio_ablation_summary.json` | measures current AAC audio contribution and a raw log-mel audio replacement across all 12 task contracts |
+| Audio ablation | `scripts/audio_ablation_and_raw_upgrade.py`, `results/audio_ablation/`, `docs/data/audio_ablation_summary.json` | measures whether audio helps each of the 12 task contracts |
 | Research roadmap | `RESEARCH_ROADMAP.md`, `docs/research_roadmap.html`, `docs/data/research_roadmap.json`, `docs/data/research_roadmap_interactive.json` | stages and visualizes the path from public-sample task development to multi-episode held-out evaluation, foundation-model selection, and larger omni/world-model extensions |
 | Foundation-model plan | `FOUNDATION_MODEL_PLAN.md`, `docs/data/foundation_model_plan.json` | keeps Qwen3-Omni as the first trainable pilot, adds Cosmos 3 as the first world-model branch, and tracks OpenVLA/openpi/GR00T policy candidates |
 | 12-task suite | `scripts/episode_task_suite.py`, per-task `metrics.json`, predictions | chronological single-episode split |
@@ -106,12 +103,11 @@ multi-episode held-out model metrics:
 | Task surface integrity | `docs/data/task_surface_integrity.json`, `scripts/validate_task_surface.py` | public task cards stay human-readable, thumbnail-backed, and wired to the scrub/play walkthrough storyboard |
 | Rendered website check | `RENDERED_SITE_CHECK.md`, `docs/data/rendered_site_check.json`, `scripts/build_rendered_site_check.py` | records a browser-level load, tab, walkthrough deep-link, control-click, and console-health check |
 | Public project surface | `PUBLIC_SURFACE_QA.md`, `docs/data/public_surface_qa.json`, `scripts/build_public_surface_qa.py` | presents the repo, website, and Hugging Face cards as one research project surface |
-| Qwen3-Omni | `results/omni_finetune/DATA_ACCESS_STATUS.md`, `MULTI_EPISODE_ACCESS_STATUS.md` | setup-stage until 32 valid episodes are available and held-out evaluation runs |
-| Multi-episode pilot status | `scripts/validate_scope_claims.py`, `docs/data/scope_claims_audit.json` | records setup-stage `32ep` artifacts separately from completed held-out-episode metrics |
+| Qwen3-Omni | `results/omni_finetune/DATA_ACCESS_STATUS.md`, `MULTI_EPISODE_ACCESS_STATUS.md` | full-dataset access is granted; 128 selected episodes are in accelerated relay/staging before held-out evaluation |
+| Multi-episode pilot status | `scripts/validate_scope_claims.py`, `docs/data/scope_claims_audit.json` | separates setup artifacts, selected relay state, and completed held-out-episode metrics |
 | Mirror parity | `scripts/validate_mirror_parity.py`, `docs/data/mirror_parity.json` | prepared GitHub/HF mirrors carry matching data, figure, website HTML, and validator files |
 | Public bundle contents | `scripts/validate_publication_package.py`, `docs/data/publication_audit.json` | summarizes the public repo and HF bundles, including raw-data exclusion and local scratch-file exclusion |
 | Release checks | `QUALITY_GATES.md`, `docs/data/quality_gates.json`, `scripts/build_quality_gates.py` | one map for automated checks and live post-publish verification |
-| Model mirror metrics | `metrics/quality_gates.json`, `metrics/public_surface_qa.json`, `metrics/mirror_parity.json` | model-repo copies of the release checks and public-surface reports |
 | Artifact index | `scripts/build_artifact_index.py`, `docs/data/artifact_index.json` | selective source-of-truth catalog with existence, size, and stable-file hashes |
 | Project status | `PROJECT_STATUS.md`, `docs/data/project_status.json` | compact current-state table for first-pass readers |
 | Citation and metadata | `CITATION.cff`, `codemeta.json`, `docs/data/project_manifest.json`, `LICENSE` | code is MIT-scoped; raw-data use follows Xperience-10M terms |
@@ -182,14 +178,14 @@ They give the current research state in one compact table:
 
 | Area | Current decision |
 | --- | --- |
-| Public-sample pipeline | Verified on one public sample episode: 5,821 frames, 1,161 windows, 8,546 current features |
+| Public-sample pipeline | Verified on one public sample episode: 5,821 frames, 1,161 windows, 8,546 dimensions |
 | 12-task suite | Verified minimal baselines with committed metrics, predictions, and manifests |
 | Neural heads | Verified compact PyTorch MLP heads over the same task contracts and chronological splits |
 | Official dataset wording | Verified against the public `ropedia-ai/xperience-10m` dataset card/API metadata |
 | Source alignment | Source facts, sample details, API-listing notes, and project coverage are consistent across repo, website, and HF cards |
 | Evaluation protocol | Verified generated protocol for windowing, split policy, leakage controls, and per-task metrics |
 | Website and HF mirrors | Verified by website reference reports, public project-surface reports, mirror parity, and live-publication checks; the public dashboard uses five top-level tabs plus subsection tabs for dataset, task-suite, method, result, and resource views |
-| Qwen3-Omni multi-episode pilot | Full-dataset access granted; 128-episode relay in progress, with full metrics pending completed staging and held-out evaluation |
+| Qwen3-Omni multi-episode pilot | Full-dataset access granted; 128-episode relay in progress with chunked parallel transfer and batch prefetch, with full metrics pending completed staging and held-out evaluation |
 | Raw Xperience-10M data / full Qwen weights | Not redistributed |
 
 ## 90-Second Research Project Path
@@ -206,7 +202,7 @@ If you are reading the project cold, open these in order:
 | 6 | What is the staged roadmap? | [`RESEARCH_ROADMAP.md`](RESEARCH_ROADMAP.md), [`docs/data/research_roadmap.json`](docs/data/research_roadmap.json), [`DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md) | The roadmap connects public-sample task development to multi-episode staging, Qwen3-Omni LoRA, foundation-model selection, robustness runs, and larger omni/world-model extensions. |
 | 7 | Which foundation model comes next? | [`FOUNDATION_MODEL_PLAN.md`](FOUNDATION_MODEL_PLAN.md), [`docs/data/foundation_model_plan.json`](docs/data/foundation_model_plan.json) | Qwen3-Omni remains the first held-out LoRA baseline; Cosmos 3 is the first world-model branch; OpenVLA/openpi/GR00T wait for explicit action targets. |
 | 8 | How do I reproduce it? | [`REPRODUCIBILITY.md`](REPRODUCIBILITY.md), [`docs/data/reproducibility_matrix.json`](docs/data/reproducibility_matrix.json), [`notes/reproducibility_audit.md`](notes/reproducibility_audit.md) | Public commands, expected outputs, and the latest exact-match reproduction record are explicit. |
-| 9 | What is one model input? | [`windows.csv`](results/episode_task_suite/windows.csv), [`feature_manifest.json`](results/episode_task_suite/feature_manifest.json), [`available_modalities.json`](results/episode_task_suite/available_modalities.json) | The input is an aligned 8,546-d window vector with explicit feature-block boundaries. |
+| 9 | What is one model input? | [`windows.csv`](results/episode_task_suite/windows.csv), [`feature_manifest.json`](results/episode_task_suite/feature_manifest.json), [`available_modalities.json`](results/episode_task_suite/available_modalities.json) | The input is an aligned 8,546-dimensional multimodal window with synchronized video, audio, sensor, and language signals. |
 | 10 | Are the task results backed by files? | [`summary_report.json`](results/episode_task_suite/summary_report.json), [`neural_mlp/`](results/episode_task_suite/neural_mlp/), [`docs/data/summary_metrics.json`](docs/data/summary_metrics.json) | Each task has minimal and neural-head evidence over the same window contracts. |
 | 11 | Is the website self-consistent? | [`docs/data/website_integrity.json`](docs/data/website_integrity.json), [`scripts/validate_website_integrity.py`](scripts/validate_website_integrity.py) | Local links, anchors, tab routing, JSON data, and referenced images are checked before publishing. |
 | 12 | What is still pending? | [`DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md), [`MULTI_EPISODE_ACCESS_STATUS.md`](results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md), [`scripts/omni/discover_xperience10m_sources.py`](scripts/omni/discover_xperience10m_sources.py) | The multi-episode Qwen3-Omni run is prepared at the selection and relay level; final model metrics require completed staging, preprocessing, training, and held-out evaluation. |
@@ -235,7 +231,7 @@ generated from committed metric artifacts. They define:
 - the 20-frame window unit, stride, feature dimension, and raw-data policy,
 - the chronological 70/30 single-episode split and its generalization limit,
 - the per-task input, target, primary metric, minimal score, and neural score,
-- leakage controls for future labels, target feature blocks, caption/object
+- leakage controls for future labels, target-side signals, caption/object
   labels, and train-only normalization,
 - current limitations, including cross-episode generalization,
   audio-visual learning, pixel-depth reconstruction, and real held-out
@@ -278,11 +274,11 @@ the selected 128-episode held-out multi-episode relay now in progress.
 This repo's current verified subset is much smaller and intentionally explicit:
 
 - one public sample episode, 5,821 frames, and 1,161 aligned windows,
-- raw sample files with six MP4 video streams and AAC audio streams,
+- raw sample files with six MP4 video streams and audio streams,
 - `annotation.hdf5` carrying depth, SLAM/camera pose, hand/body mocap, IMU,
   language/caption annotations, calibration, metadata, and timing records,
-- an 8,546-d baseline feature vector using video-derived statistics, AAC audio,
-  depth, pose/SLAM, mocap, IMU, calibration, and language-derived blocks.
+- an 8,546-dimensional baseline representation using video, audio, depth,
+  pose/SLAM, mocap, IMU, calibration, and language-derived signals.
 
 The same alignment note also records what is outside the current implemented subset: real
 audio-visual learning, caption generation, pixel-depth estimation, SLAM
@@ -589,7 +585,7 @@ python scripts/omni/plan_finetune_sample_budget.py \
   --full-preview-per-episode-gb 5.1
 ```
 
-### 32-Episode Readiness Gate
+### Multi-Episode Readiness Gate
 
 ```bash
 python scripts/omni/discover_xperience10m_sources.py \
@@ -604,8 +600,8 @@ Current status in this repo:
 - gated_metadata_audit: 12,102 complete visible episodes across 802 complete sessions
 - selected_relay_plan: 128 metadata-balanced episodes, 96/16/16 train/val/test
 - selected_download_size: 277.71 GiB excluding `visualization.rrd`
-- ready_for_held_out_pilot: false until the selected episodes are fully staged and audited
-- full-dataset access: granted; raw multi-episode staging is in progress
+- ready_for_held_out_pilot: false until the selected episodes are fully staged and checked
+- full-dataset access: granted; raw multi-episode staging is in progress with chunked parallel transfer and overlapping batch prefetch
 - source_discovery: `results/omni_finetune/source_discovery.json`
 - data_status: `results/omni_finetune/DATA_ACCESS_STATUS.md`
 - access_status: `results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`
@@ -662,7 +658,7 @@ The taxonomy uses two current baselines for every task:
 
 | Baseline | Role |
 | --- | --- |
-| Minimal interpretable heads | Softmax, logistic, ridge, and retrieval heads over the 8,546-d window feature vector. These expose the input/output contract cleanly. |
+| Minimal interpretable heads | Softmax, logistic, ridge, and retrieval heads over the 8,546-dimensional multimodal representation. These expose the input/output contract cleanly. |
 | Neural MLP heads | Small PyTorch MLP classifiers/regressors on the same features and splits. These check whether nonlinear heads help before moving to Qwen/Omni fine-tuning. |
 
 Current direction-level coverage:
@@ -755,7 +751,7 @@ models.
 Shared setup:
 
 ```text
-raw episode -> 20-frame windows, stride 5 -> 8,546-d current feature vector
+raw episode -> 20-frame windows, stride 5 -> 8,546-dimensional multimodal representation
 chronological split: first 70% train, last 30% test
 scalers are fit on train windows only
 ```
@@ -769,7 +765,7 @@ There are four reusable head families:
 | Ridge + cosine ranking | Language Grounding, Cross-Modal Retrieval | project one modality into another feature space, then rank candidates by cosine |
 | Multi-label logistic regression | Object Relevance Prediction | z-score non-caption features, sigmoid object heads, threshold at 0.5 |
 
-The optional neural run keeps the same feature vectors, leakage filters,
+The optional neural run keeps the same window representation, leakage filters,
 chronological splits, and metrics, but replaces the task heads with small
 PyTorch MLP classifiers or regressors. Its outputs live under
 [`results/episode_task_suite/neural_mlp/`](results/episode_task_suite/neural_mlp/),
@@ -785,11 +781,11 @@ The task-specific heads are:
 | Action Boundary Detection | all featurized modalities | linear softmax | steady vs action boundary |
 | Next-Action Prediction | all featurized modalities at `t` | linear softmax | action at `t+20` frames |
 | Hand Trajectory Forecasting | all featurized modalities at `t` | ridge regression | future 10-frame left/right hand joints |
-| Contact State Prediction | non-contact and non-caption feature blocks | linear softmax | any body contact |
-| Object Relevance Prediction | non-caption feature blocks | multi-label logistic | relevant object set |
+| Contact State Prediction | non-contact and non-caption signals | linear softmax | any body contact |
+| Object Relevance Prediction | non-caption signals | multi-label logistic | relevant object set |
 | Language Grounding | sensor windows projected to text space | ridge projection + cosine ranking | matching time window for text query |
 | Cross-Modal Retrieval | motion/IMU/camera projected to visual space | ridge projection + cosine ranking | matching depth/video window |
-| Cross-Modal Reconstruction | motion/IMU/camera | ridge regression | depth/video feature vector |
+| Cross-Modal Reconstruction | motion/IMU/camera | ridge regression | compressed depth/video target |
 | Temporal Order Verification | `[x_t, x_t+1, x_t+1-x_t]` | binary linear softmax | correct vs reversed order |
 | Multimodal Synchronization Detection | motion plus visual pair | binary linear softmax | aligned vs shifted by 8 windows |
 
@@ -798,7 +794,7 @@ The task-specific heads are:
 | Experiment | Main score | Accuracy | Notes |
 | --- | ---: | ---: | --- |
 | Motion-only action | 0.9688 macro-F1 | 0.9828 | Uses motion/IMU features only |
-| Current all-feature action | 0.9829 macro-F1 | 0.9863 | 8,546-dimensional feature vector |
+| Current all-feature action | 0.9829 macro-F1 | 0.9863 | 8,546-dimensional multimodal representation |
 | Motion-only subtask | 0.9528 macro-F1 | 0.9759 | Strong within-episode subtask signal |
 | Current all-feature subtask | 0.9173 macro-F1 | 0.9828 | High accuracy, lower class-balanced score |
 | Cross-modal retrieval | 0.3678 top-5 | n/a | Motion/IMU/camera/audio retrieves matching depth/video |
@@ -807,27 +803,26 @@ The task-specific heads are:
 | Neural MLP hand forecast | 0.1079 MPJPE | n/a | Same features/split, nonlinear regression head |
 | Neural MLP temporal order | 0.8520 F1 | 0.8578 | Strong improvement on adjacent-window ordering |
 | Neural MLP misalignment | 0.7153 F1 | 0.7009 | Detects shifted motion/visual/audio pairs better than the linear head |
-| Audio ablation | +0.0418 mean delta | n/a | Current AAC audio improves the primary metric on 6 of 12 task contracts |
-| Raw log-mel audio replacement | +0.0936 mean delta | n/a | Raw log-mel replacement beats current handcrafted audio on 6 of 12 task contracts |
+| Audio ablation | +0.0418 mean delta | n/a | Current audio variant improves the primary metric on 6 of 12 task contracts |
+| Alternate audio representation | +0.0936 mean delta | n/a | Alternate audio-window representation improves over the baseline audio variant on 6 of 12 task contracts |
 
-## Audio Ablation and Raw-Audio Upgrade
+## Audio Contribution Study
 
-The current AAC audio block is now tested rather than only included. The script
+The audio ablation keeps the same windows and task labels, then compares input
+variants under the same chronological split. The script
 [`scripts/audio_ablation_and_raw_upgrade.py`](scripts/audio_ablation_and_raw_upgrade.py)
-reuses the real task-suite windows, decodes the local public-sample
-`fisheye_cam0.mp4` audio stream, builds a 588-d raw log-mel window feature, and
-evaluates six variants for every task: current features, no audio,
-handcrafted-audio-only, raw-audio-only, handcrafted audio replaced by raw
-log-mel, and current features plus raw log-mel.
+reuses the real task-suite windows and evaluates six variants for
+every task: current inputs, no audio, audio-only, alternate audio-only, audio
+representation replacement, and all inputs plus the alternate audio representation.
 
 The measured single-episode result is task-specific:
 
 | Readout | Value |
 | --- | ---: |
-| Tasks where current AAC audio improves the primary metric | 6 / 12 |
+| Tasks where current audio improves the primary metric | 6 / 12 |
 | Mean current-audio delta | +0.0418 |
-| Tasks where raw log-mel replacement improves over handcrafted AAC | 6 / 12 |
-| Mean raw-replacement delta vs current audio | +0.0936 |
+| Tasks where alternate audio representation improves over baseline audio | 6 / 12 |
+| Mean alternate-representation delta vs baseline audio | +0.0936 |
 
 Full files:
 
@@ -842,7 +837,7 @@ Full files:
 The neural baseline was run locally with `--include-neural` for all 12 tasks
 using 80 epochs, hidden size 128, batch size 128, and CPU execution. It is not a
 foundation model result; it is a controlled nonlinear-head comparison over the
-same 8,546-d handcrafted window features.
+same 8,546-dimensional multimodal representation.
 
 | Task | Neural metric | Minimal metric | Readout |
 | --- | ---: | ---: | --- |
@@ -877,10 +872,10 @@ artifact-driven diagnostics pass over the public sample episode:
 - `results/single_episode_diagnostics/alignment_stress/alignment_shift_metrics.csv`
   evaluates cross-modal retrieval under explicit time shifts.
 - `docs/single_episode_explorer.html` is a static interactive page for
-  inspecting window labels, objects, predictions, feature-block statistics, and
+  inspecting window labels, objects, predictions, modality statistics, and
   diagnostic scores.
 
-These are single-episode research diagnostics. They are useful for auditing
+These are single-episode research diagnostics. They are useful for studying
 task definitions, feature behavior, and model errors before scaling to more
 episodes; they are not reported as multi-episode benchmark results.
 
@@ -907,21 +902,21 @@ The test segment contains some action/subtask labels never seen during training.
 Timeline and next-action classifiers therefore expose the core limitation of
 single-episode learning instead of hiding it behind random splits.
 
-## Feature Blocks Used
+## Modalities Used
 
-The current feature vector has 8,546 dimensions and includes:
+The current public-sample pipeline uses:
 
 - hand/body mocap joints and contact labels,
 - camera translation and rotation,
 - IMU acceleration and gyroscope traces,
 - depth confidence features,
 - six video streams,
-- AAC audio features from `fisheye_cam0.mp4`,
+- audio from the sample MP4 stream,
 - caption/object/interaction text features,
 - SLAM point-cloud summary features,
 - calibration parameters.
 
-The exact feature block boundaries are stored in
+The full technical source manifest is stored in
 [`results/episode_task_suite/feature_manifest.json`](results/episode_task_suite/feature_manifest.json).
 
 ## Data Notice

REPRODUCIBILITY.md

@@ -10,11 +10,11 @@ outside the current public data scope.
 | --- | --- | --- |
 | Sample download | Yes, from `ropedia-ai/xperience-10m-sample` or ModelScope sample mirror | Sample card lists `cc-by-nc-4.0`; raw data is not redistributed in this repo. |
 | Minimal baselines | Yes | One public sample episode, chronological split. |
-| 12-task suite | Yes | Uses the current 8,546-d feature contract, including the decoded AAC audio block. |
+| 12-task suite | Yes | Uses the current 8,546-d synchronized multimodal feature contract. |
 | Neural MLP heads | Yes, when `torch` is installed | Compact task heads only, not a foundation model. |
 | Website figures and charts | Yes | Generated from committed metrics and sample thumbnails. |
 | Public bundle contents | Yes | Covers public repo and prepared HF bundles. |
-| 32-episode Qwen3-Omni LoRA pilot | Not yet | Gated by full Xperience-10M access and held-out-episode evaluation. |
+| Multi-episode Qwen3-Omni LoRA pilot | Not yet | Full-dataset access is granted; held-out metrics require completed staging, training, and evaluation. |
 
 ## Environment
 
@@ -129,7 +129,7 @@ Evidence:
 The following require gated data, large model weights, or private compute
 state, so this repo does not yet provide public reproduction for:
 
-- a real 32-episode Qwen3-Omni LoRA run,
+- a real held-out multi-episode Qwen3-Omni LoRA run,
 - held-out episode metrics for Qwen3-Omni,
 - full Xperience-10M-scale pretraining,
 - raw Xperience-10M video or annotation redistribution,

RESEARCH_ROADMAP.md

@@ -10,17 +10,17 @@ should exist before the stage is treated as complete.
 | Stage | Status | Entry condition | Research deliverables | Completion evidence |
 | --- | --- | --- | --- | --- |
 | Public-Sample Task Lab | Implemented | One public Xperience-10M sample episode is available. | 1,161 aligned windows, 12 task contracts, minimal heads, neural MLP heads, modality atlas, task walkthroughs, and derived figures. | `PROJECT_STATUS.md`, `EVALUATION_PROTOCOL.md`, `RESEARCH_TAKEAWAYS.md`, `docs/data/summary_metrics.json`, `results/episode_task_suite/summary_report.json` |
-| Multi-Episode Data Staging | Active | Gated dataset access and enough storage for selected episodes. | 32 valid episodes, episode manifest, missing-view manifest, held-out episode split, and source-discovery report. | `results/omni_finetune/DATA_ACCESS_STATUS.md`, `results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`, `results/omni_finetune/source_discovery.json` |
-| 32-Episode Qwen3-Omni LoRA Pilot | Next | At least 32 valid episodes staged locally with no train/test episode leakage. | Dataset JSONL/media manifests, LoRA adapter checkpoint, progress logs, held-out predictions, metrics, confusion matrices, and run report. | `dataset_manifest.json`, `training_metadata.json`, `progress.jsonl`, `metrics.json`, `predictions.jsonl`, `RUN_REPORT.md` |
-| Foundation-Model Selection Matrix | Next | 32-episode data gate is satisfied, or a 3-8 episode dry run is staged for preprocessing checks. | Backbone registry, Cosmos 3 world-model branch plan, Qwen3-Omni baseline plan, OpenVLA/openpi/GR00T policy candidates, and model-specific evaluation additions. | `FOUNDATION_MODEL_PLAN.md`, `docs/data/foundation_model_plan.json`, `research_roadmap_interactive.json` |
-| 64-128 Episode Robustness Run | Planned | The 32-episode pilot trains and evaluates cleanly. | Split-by-session metrics, modality ablations, calibration/object/language error analysis, and sensitivity to missing views. | Held-out metrics by session, task, and modality; ablation tables; qualitative error analysis. |
+| Multi-Episode Data Staging | Active | Full-dataset access and enough storage for selected episodes. | 128 selected episodes, episode manifest, missing-view manifest, held-out episode split, and source-discovery report. | `results/omni_finetune/DATA_ACCESS_STATUS.md`, `results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`, `results/omni_finetune/source_discovery.json` |
+| Qwen3-Omni LoRA Pilot | Next | Selected episodes staged locally with no train/test episode leakage. | Dataset JSONL/media manifests, LoRA adapter checkpoint, progress logs, held-out predictions, metrics, confusion matrices, and run report. | `dataset_manifest.json`, `training_metadata.json`, `progress.jsonl`, `metrics.json`, `predictions.jsonl`, `RUN_REPORT.md` |
+| Foundation-Model Selection Matrix | Next | The selected relay is staged, or a 3-8 episode dry run is staged for preprocessing checks. | Backbone registry, Cosmos 3 world-model branch plan, Qwen3-Omni baseline plan, OpenVLA/openpi/GR00T policy candidates, and model-specific evaluation additions. | `FOUNDATION_MODEL_PLAN.md`, `docs/data/foundation_model_plan.json`, `research_roadmap_interactive.json` |
+| 64-128 Episode Robustness Run | Planned | The selected-episode pilot trains and evaluates cleanly. | Split-by-session metrics, modality ablations, calibration/object/language error analysis, and sensitivity to missing views. | Held-out metrics by session, task, and modality; ablation tables; qualitative error analysis. |
 | Cosmos 3 and Policy-Model Extensions | Planned | Enough multi-episode data, compute budget, and model-specific action/world-state targets. | Cosmos 3 future-window or action-conditioned world-model probes, OpenVLA/openpi/GR00T action-policy baselines, modality-conditioning audits, affordance tasks, and synthetic-data usefulness tests. | Task-specific held-out evaluations, qualitative inspection, and updated model cards. |
 
 ## Current Decision Point
 
 The useful next decision is data scale plus backbone fit: keep the public-sample
 task suite as the development harness, stage enough official Xperience-10M
-episodes to run the 32-episode held-out pilot, then choose larger model branches
+episodes to run the held-out Qwen3-Omni pilot, then choose larger model branches
 by task fit. Qwen3-Omni remains the first trainable multimodal LoRA target.
 Cosmos 3 becomes the first world-model/action-generation branch. OpenVLA,
 openpi, GR00T, Octo, and SmolVLA-style models become policy/action branches only
@@ -58,7 +58,7 @@ Evidence to inspect:
 - `scripts/omni/discover_xperience10m_sources.py`
 - `results/omni_finetune/source_discovery.json`
 
-### 3. 32-Episode Qwen3-Omni LoRA Pilot
+### 3. Qwen3-Omni LoRA Pilot
 
 This stage uses Qwen3-Omni as the multimodal backbone and trains lightweight
 LoRA adapters. The first target is a complete held-out-episode training and
@@ -78,7 +78,7 @@ Expected outputs:
 
 ### 4. 64-128 Episode Robustness Run
 
-This stage asks whether the 32-episode conclusions survive more sessions,
+This stage asks whether the pilot conclusions survive more sessions,
 different objects, missing views, and stronger modality ablations. It should
 report performance by task, session, modality, and failure type.
 

RESEARCH_TAKEAWAYS.md

@@ -11,7 +11,7 @@ from hand-edited score text.
 - aligned windows: 1,161
 - current feature dimension: 8,546
 - raw Xperience-10M data is not redistributed
-- AAC audio from the sample MP4 stream is extracted into the current feature vector
+- Audio from the sample MP4 stream is represented in the current feature vector
 
 ## Takeaways
 
@@ -80,7 +80,7 @@ Current scope: The current reconstruction task predicts feature vectors; depth,
 
 ### Audio helps some tasks and hurts others on the public sample
 
-The current AAC audio block improves the primary metric on 6 of 12 tasks, while raw log-mel replacement improves over the current handcrafted block on 6 of 12 tasks. The largest current-audio gain appears in feature reconstruction, not in action classification.
+Audio improves the primary metric on 6 of 12 tasks, while raw log-mel replacement improves over the current handcrafted block on 6 of 12 tasks. The largest current-audio gain appears in feature reconstruction, not in action classification.
 
 | Metric | Value |
 | --- | ---: |
@@ -97,23 +97,23 @@ Current scope: This is a single-episode ablation over fixed ridge heads. It vali
 
 ### The next scientific unit is held-out episodes, not more adjacent windows
 
-The prepared Qwen3-Omni path targets 32 episodes from 32 sessions, but it remains data-gated until access and held-out evaluation complete.
+The prepared Qwen3-Omni path now targets a selected 128-episode pilot; held-out metrics will be reported after staging, training, and evaluation complete.
 
 | Metric | Value |
 | --- | ---: |
-| `target_episodes` | 32 |
-| `selected_sessions` | 32 |
-| `valid_candidates` | 680 |
+| `target_episodes` | 128 |
+| `selected_sessions` | 128 |
+| `valid_candidates` | 12,102 |
 
 Source: `results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`.
 
-Current scope: The 32-episode Qwen3-Omni fine-tune requires gated data staging and held-out evaluation.
+Current scope: The selected-episode Qwen3-Omni fine-tune requires completed data staging and held-out evaluation.
 
 ## How To Read These Results
 
 - High single-episode scores are useful pipeline checks for the current task contracts.
 - Low chronological action/subtask scores are informative because they expose later-label shift.
 - Neural gains on trajectory/order/alignment make those tasks good candidates for the next fine-tuning stage.
-- Audio ablation is task-specific: current AAC and raw log-mel features help some probes and hurt others.
+- Audio ablation is task-specific: audio representation choices help some probes and hurt others.
 - Retrieval and reconstruction remain the main multimodal representation challenges.
 - The next credible model-quality result needs held-out episodes.

XPERIENCE10M_DATASET_CARD_ALIGNMENT.md

@@ -156,9 +156,8 @@ multimodal records. The relevant groups include:
 - language/caption annotations and metadata
 
 This repo's current 8,546-d feature vector uses video-derived statistics,
-AAC audio, depth, pose/SLAM, calibration, mocap, IMU, and language-derived
-blocks. The AAC block is decoded from `fisheye_cam0.mp4` and recorded in the
-feature manifest as `audio_fisheye_cam0_aac`.
+audio, depth, pose/SLAM, calibration, mocap, IMU, and language-derived
+blocks.
 
 ## Intended Research Uses
 
@@ -209,7 +208,7 @@ When describing Xperience-10M in this repo, keep these limitations visible:
 - motion capture, SLAM, depth, captions, and other annotations can contain noise
 - language annotations are not exhaustive descriptions of every scene state
 - large-scale training requires substantial storage, preprocessing, and compute
-- the current feature vector includes a compact AAC audio feature block, while
+- the current feature vector includes compact audio features, while
   larger audio-visual representation learning remains a multi-episode milestone
 
 ## Current Project Alignment
@@ -221,6 +220,6 @@ When describing Xperience-10M in this repo, keep these limitations visible:
 | Public sample repo is `cc-by-nc-4.0` and points to HOMIE/Rerun | Preserved in data notice and reproducibility docs |
 | Public sample includes video/audio/depth/pose/mocap/IMU/language | Represented in the modality atlas |
 | Episode layout uses six MP4 streams and `annotation.hdf5` | Used by sample inspection and pilot-readiness scripts |
-| Audio exists in MP4 streams | Decoded into the current `audio_fisheye_cam0_aac` feature block |
+| Audio exists in MP4 streams | Represented in the current multimodal feature contract |
 | 4D reconstruction/world modeling are intended research directions | Represented by proxy/diagnostic tasks only |
 | Real model quality requires held-out multi-episode evaluation | Pending selected multi-episode staging, training, and held-out evaluation |

data/mirror_parity.json

@@ -1,9 +1,9 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-03T07:52:34+00:00",
+  "generated_at_utc": "2026-06-03T19:30:53+00:00",
   "hf_root": "hf_publish",
   "summary": {
-    "group_count": 91,
+    "group_count": 101,
     "failure_count": 0,
     "failures_by_surface": {}
   },
@@ -34,33 +34,64 @@
     }
   ],
   "groups": [
+    {
+      "name": "data/audio_ablation_summary.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:docs/data/audio_ablation_summary.json",
+        "exists": true,
+        "bytes": 9701,
+        "sha256": "7debd9e67b7df09322d743193bf8b785277a992c9e8c82c08c7a36c4e066e6de"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:data/audio_ablation_summary.json",
+          "exists": true,
+          "bytes": 9701,
+          "sha256": "7debd9e67b7df09322d743193bf8b785277a992c9e8c82c08c7a36c4e066e6de"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:docs/data/audio_ablation_summary.json",
+          "exists": true,
+          "bytes": 9701,
+          "sha256": "7debd9e67b7df09322d743193bf8b785277a992c9e8c82c08c7a36c4e066e6de"
+        },
+        "hf_model": {
+          "path": "hf_model:metrics/audio_ablation_summary.json",
+          "exists": true,
+          "bytes": 9701,
+          "sha256": "7debd9e67b7df09322d743193bf8b785277a992c9e8c82c08c7a36c4e066e6de"
+        }
+      },
+      "failures": []
+    },
     {
       "name": "data/artifact_index.json",
       "status": "pass",
       "local": {
         "path": "repo:docs/data/artifact_index.json",
         "exists": true,
-        "bytes": 29016,
-        "sha256": "e86a70ddecc41330f7e0549dc3d7f7af9984cd92d5b5e5f181ba29c95833b62a"
+        "bytes": 32292,
+        "sha256": "dcd253ba6636ea4bf78260b51a1493b4faf7cf5fa22e74d67d4cd277b9fe4189"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/artifact_index.json",
           "exists": true,
-          "bytes": 29016,
-          "sha256": "e86a70ddecc41330f7e0549dc3d7f7af9984cd92d5b5e5f181ba29c95833b62a"
+          "bytes": 32292,
+          "sha256": "dcd253ba6636ea4bf78260b51a1493b4faf7cf5fa22e74d67d4cd277b9fe4189"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/artifact_index.json",
           "exists": true,
-          "bytes": 29016,
-          "sha256": "e86a70ddecc41330f7e0549dc3d7f7af9984cd92d5b5e5f181ba29c95833b62a"
+          "bytes": 32292,
+          "sha256": "dcd253ba6636ea4bf78260b51a1493b4faf7cf5fa22e74d67d4cd277b9fe4189"
         },
         "hf_model": {
           "path": "hf_model:metrics/artifact_index.json",
           "exists": true,
-          "bytes": 29016,
-          "sha256": "e86a70ddecc41330f7e0549dc3d7f7af9984cd92d5b5e5f181ba29c95833b62a"
+          "bytes": 32292,
+          "sha256": "dcd253ba6636ea4bf78260b51a1493b4faf7cf5fa22e74d67d4cd277b9fe4189"
         }
       },
       "failures": []
@@ -102,27 +133,27 @@
       "local": {
         "path": "repo:docs/data/evidence_contract.json",
         "exists": true,
-        "bytes": 12007,
-        "sha256": "abf0923da5e24b2773b19ecbc83873b05ecc99be07089c63e038a354d806217c"
+        "bytes": 12096,
+        "sha256": "3d2815d31952a5102c78818a852db8ff36faf49d89643b32ad59bcda91d499cb"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/evidence_contract.json",
           "exists": true,
-          "bytes": 12007,
-          "sha256": "abf0923da5e24b2773b19ecbc83873b05ecc99be07089c63e038a354d806217c"
+          "bytes": 12096,
+          "sha256": "3d2815d31952a5102c78818a852db8ff36faf49d89643b32ad59bcda91d499cb"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/evidence_contract.json",
           "exists": true,
-          "bytes": 12007,
-          "sha256": "abf0923da5e24b2773b19ecbc83873b05ecc99be07089c63e038a354d806217c"
+          "bytes": 12096,
+          "sha256": "3d2815d31952a5102c78818a852db8ff36faf49d89643b32ad59bcda91d499cb"
         },
         "hf_model": {
           "path": "hf_model:metrics/evidence_contract.json",
           "exists": true,
-          "bytes": 12007,
-          "sha256": "abf0923da5e24b2773b19ecbc83873b05ecc99be07089c63e038a354d806217c"
+          "bytes": 12096,
+          "sha256": "3d2815d31952a5102c78818a852db8ff36faf49d89643b32ad59bcda91d499cb"
         }
       },
       "failures": []
@@ -133,27 +164,27 @@
       "local": {
         "path": "repo:docs/data/evaluation_protocol.json",
         "exists": true,
-        "bytes": 13667,
-        "sha256": "3fa79360af0f055816b50fca655455b321430c0b59d386991d3ec9776aef0bb3"
+        "bytes": 13686,
+        "sha256": "a2a35bd3b05e228c99c5ef7cc002c84708dcaff893cd01d51b011dadbf2c02c4"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/evaluation_protocol.json",
           "exists": true,
-          "bytes": 13667,
-          "sha256": "3fa79360af0f055816b50fca655455b321430c0b59d386991d3ec9776aef0bb3"
+          "bytes": 13686,
+          "sha256": "a2a35bd3b05e228c99c5ef7cc002c84708dcaff893cd01d51b011dadbf2c02c4"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/evaluation_protocol.json",
           "exists": true,
-          "bytes": 13667,
-          "sha256": "3fa79360af0f055816b50fca655455b321430c0b59d386991d3ec9776aef0bb3"
+          "bytes": 13686,
+          "sha256": "a2a35bd3b05e228c99c5ef7cc002c84708dcaff893cd01d51b011dadbf2c02c4"
         },
         "hf_model": {
           "path": "hf_model:metrics/evaluation_protocol.json",
           "exists": true,
-          "bytes": 13667,
-          "sha256": "3fa79360af0f055816b50fca655455b321430c0b59d386991d3ec9776aef0bb3"
+          "bytes": 13686,
+          "sha256": "a2a35bd3b05e228c99c5ef7cc002c84708dcaff893cd01d51b011dadbf2c02c4"
         }
       },
       "failures": []
@@ -165,26 +196,57 @@
         "path": "repo:docs/data/figure_index.json",
         "exists": true,
         "bytes": 13434,
-        "sha256": "626ce54a4486a3ce8f284ff21004c0cc485886acedd31c77d80ba115bd07f09c"
+        "sha256": "1a28548d2838b5a5ea9464761ccb73a3e55d6d48dd4c68d321f056bfdc4ad560"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/figure_index.json",
           "exists": true,
           "bytes": 13434,
-          "sha256": "626ce54a4486a3ce8f284ff21004c0cc485886acedd31c77d80ba115bd07f09c"
+          "sha256": "1a28548d2838b5a5ea9464761ccb73a3e55d6d48dd4c68d321f056bfdc4ad560"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/figure_index.json",
           "exists": true,
           "bytes": 13434,
-          "sha256": "626ce54a4486a3ce8f284ff21004c0cc485886acedd31c77d80ba115bd07f09c"
+          "sha256": "1a28548d2838b5a5ea9464761ccb73a3e55d6d48dd4c68d321f056bfdc4ad560"
         },
         "hf_model": {
           "path": "hf_model:metrics/figure_index.json",
           "exists": true,
           "bytes": 13434,
-          "sha256": "626ce54a4486a3ce8f284ff21004c0cc485886acedd31c77d80ba115bd07f09c"
+          "sha256": "1a28548d2838b5a5ea9464761ccb73a3e55d6d48dd4c68d321f056bfdc4ad560"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "data/foundation_model_plan.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:docs/data/foundation_model_plan.json",
+        "exists": true,
+        "bytes": 8880,
+        "sha256": "a11476cc697e7bc2c29b3db569d059c8f536025fdd0b47e49f060a88664fd450"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:data/foundation_model_plan.json",
+          "exists": true,
+          "bytes": 8880,
+          "sha256": "a11476cc697e7bc2c29b3db569d059c8f536025fdd0b47e49f060a88664fd450"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:docs/data/foundation_model_plan.json",
+          "exists": true,
+          "bytes": 8880,
+          "sha256": "a11476cc697e7bc2c29b3db569d059c8f536025fdd0b47e49f060a88664fd450"
+        },
+        "hf_model": {
+          "path": "hf_model:metrics/foundation_model_plan.json",
+          "exists": true,
+          "bytes": 8880,
+          "sha256": "a11476cc697e7bc2c29b3db569d059c8f536025fdd0b47e49f060a88664fd450"
         }
       },
       "failures": []
@@ -195,27 +257,27 @@
       "local": {
         "path": "repo:docs/data/live_publication_status.json",
         "exists": true,
-        "bytes": 68740,
-        "sha256": "d16a49a3ba54fba7327e83c40a7f1a636ab218d873737b35d37c5da77452e592"
+        "bytes": 68749,
+        "sha256": "1095bd7aeee4172afddbe7b9e89c5371605ef22fab294b3d77ec98d5ad7ef0cd"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/live_publication_status.json",
           "exists": true,
-          "bytes": 68740,
-          "sha256": "d16a49a3ba54fba7327e83c40a7f1a636ab218d873737b35d37c5da77452e592"
+          "bytes": 68749,
+          "sha256": "1095bd7aeee4172afddbe7b9e89c5371605ef22fab294b3d77ec98d5ad7ef0cd"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/live_publication_status.json",
           "exists": true,
-          "bytes": 68740,
-          "sha256": "d16a49a3ba54fba7327e83c40a7f1a636ab218d873737b35d37c5da77452e592"
+          "bytes": 68749,
+          "sha256": "1095bd7aeee4172afddbe7b9e89c5371605ef22fab294b3d77ec98d5ad7ef0cd"
         },
         "hf_model": {
           "path": "hf_model:metrics/live_publication_status.json",
           "exists": true,
-          "bytes": 68740,
-          "sha256": "d16a49a3ba54fba7327e83c40a7f1a636ab218d873737b35d37c5da77452e592"
+          "bytes": 68749,
+          "sha256": "1095bd7aeee4172afddbe7b9e89c5371605ef22fab294b3d77ec98d5ad7ef0cd"
         }
       },
       "failures": []
@@ -226,27 +288,27 @@
       "local": {
         "path": "repo:docs/data/modality_atlas.json",
         "exists": true,
-        "bytes": 3841,
-        "sha256": "8b290ea80a7fe494955fa715c658105d678c3f7f26c8379a81223aa8e67ecc24"
+        "bytes": 3797,
+        "sha256": "4f12a5471d4fe53b1f5c1ef6f9535cc14974690fa6ffbfbeddc8e1b90f86b281"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/modality_atlas.json",
           "exists": true,
-          "bytes": 3841,
-          "sha256": "8b290ea80a7fe494955fa715c658105d678c3f7f26c8379a81223aa8e67ecc24"
+          "bytes": 3797,
+          "sha256": "4f12a5471d4fe53b1f5c1ef6f9535cc14974690fa6ffbfbeddc8e1b90f86b281"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/modality_atlas.json",
           "exists": true,
-          "bytes": 3841,
-          "sha256": "8b290ea80a7fe494955fa715c658105d678c3f7f26c8379a81223aa8e67ecc24"
+          "bytes": 3797,
+          "sha256": "4f12a5471d4fe53b1f5c1ef6f9535cc14974690fa6ffbfbeddc8e1b90f86b281"
         },
         "hf_model": {
           "path": "hf_model:metrics/modality_atlas.json",
           "exists": true,
-          "bytes": 3841,
-          "sha256": "8b290ea80a7fe494955fa715c658105d678c3f7f26c8379a81223aa8e67ecc24"
+          "bytes": 3797,
+          "sha256": "4f12a5471d4fe53b1f5c1ef6f9535cc14974690fa6ffbfbeddc8e1b90f86b281"
         }
       },
       "failures": []
@@ -257,27 +319,27 @@
       "local": {
         "path": "repo:docs/data/project_brief.json",
         "exists": true,
-        "bytes": 2596,
-        "sha256": "aed77943c5cdf2557bce3ecbf75e51d5774a2419b233c601bd815c0875f447b5"
+        "bytes": 2585,
+        "sha256": "f0e2e7853b86cdcd2f06269d75f19a96432a782efdf99c08aca7d689fd2cd0a1"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/project_brief.json",
           "exists": true,
-          "bytes": 2596,
-          "sha256": "aed77943c5cdf2557bce3ecbf75e51d5774a2419b233c601bd815c0875f447b5"
+          "bytes": 2585,
+          "sha256": "f0e2e7853b86cdcd2f06269d75f19a96432a782efdf99c08aca7d689fd2cd0a1"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/project_brief.json",
           "exists": true,
-          "bytes": 2596,
-          "sha256": "aed77943c5cdf2557bce3ecbf75e51d5774a2419b233c601bd815c0875f447b5"
+          "bytes": 2585,
+          "sha256": "f0e2e7853b86cdcd2f06269d75f19a96432a782efdf99c08aca7d689fd2cd0a1"
         },
         "hf_model": {
           "path": "hf_model:metrics/project_brief.json",
           "exists": true,
-          "bytes": 2596,
-          "sha256": "aed77943c5cdf2557bce3ecbf75e51d5774a2419b233c601bd815c0875f447b5"
+          "bytes": 2585,
+          "sha256": "f0e2e7853b86cdcd2f06269d75f19a96432a782efdf99c08aca7d689fd2cd0a1"
         }
       },
       "failures": []
@@ -288,27 +350,27 @@
       "local": {
         "path": "repo:docs/data/project_manifest.json",
         "exists": true,
-        "bytes": 4644,
-        "sha256": "cf5f477e09a2cdc45b3c44078ab260df532ddd9cb64725444447f8e8f7e8f99a"
+        "bytes": 4643,
+        "sha256": "b02ca0064bb6eea07abdcf4ff4aacde06a291f94679b79ca599052848a269b49"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/project_manifest.json",
           "exists": true,
-          "bytes": 4644,
-          "sha256": "cf5f477e09a2cdc45b3c44078ab260df532ddd9cb64725444447f8e8f7e8f99a"
+          "bytes": 4643,
+          "sha256": "b02ca0064bb6eea07abdcf4ff4aacde06a291f94679b79ca599052848a269b49"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/project_manifest.json",
           "exists": true,
-          "bytes": 4644,
-          "sha256": "cf5f477e09a2cdc45b3c44078ab260df532ddd9cb64725444447f8e8f7e8f99a"
+          "bytes": 4643,
+          "sha256": "b02ca0064bb6eea07abdcf4ff4aacde06a291f94679b79ca599052848a269b49"
         },
         "hf_model": {
           "path": "hf_model:metrics/project_manifest.json",
           "exists": true,
-          "bytes": 4644,
-          "sha256": "cf5f477e09a2cdc45b3c44078ab260df532ddd9cb64725444447f8e8f7e8f99a"
+          "bytes": 4643,
+          "sha256": "b02ca0064bb6eea07abdcf4ff4aacde06a291f94679b79ca599052848a269b49"
         }
       },
       "failures": []
@@ -319,27 +381,27 @@
       "local": {
         "path": "repo:docs/data/project_packet.json",
         "exists": true,
-        "bytes": 7654,
-        "sha256": "ebf621b365c7f384ef89735b5b8829a75e86211245ba10c719cae149b9b07806"
+        "bytes": 7627,
+        "sha256": "9b36df95df2a391b93282bbf465dca1594fe3edac96bad128aa331b0b835921a"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/project_packet.json",
           "exists": true,
-          "bytes": 7654,
-          "sha256": "ebf621b365c7f384ef89735b5b8829a75e86211245ba10c719cae149b9b07806"
+          "bytes": 7627,
+          "sha256": "9b36df95df2a391b93282bbf465dca1594fe3edac96bad128aa331b0b835921a"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/project_packet.json",
           "exists": true,
-          "bytes": 7654,
-          "sha256": "ebf621b365c7f384ef89735b5b8829a75e86211245ba10c719cae149b9b07806"
+          "bytes": 7627,
+          "sha256": "9b36df95df2a391b93282bbf465dca1594fe3edac96bad128aa331b0b835921a"
         },
         "hf_model": {
           "path": "hf_model:metrics/project_packet.json",
           "exists": true,
-          "bytes": 7654,
-          "sha256": "ebf621b365c7f384ef89735b5b8829a75e86211245ba10c719cae149b9b07806"
+          "bytes": 7627,
+          "sha256": "9b36df95df2a391b93282bbf465dca1594fe3edac96bad128aa331b0b835921a"
         }
       },
       "failures": []
@@ -350,27 +412,27 @@
       "local": {
         "path": "repo:docs/data/project_status.json",
         "exists": true,
-        "bytes": 7344,
-        "sha256": "d21fb26e1accaf0130f8ae0275526a49af52a6428b2fb0de7f8dc80d4efed230"
+        "bytes": 8930,
+        "sha256": "b4f0fd7644009ff4baeb98c05beb6b8bca86b166fdb6a7dc4f263a61a7ca13dd"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/project_status.json",
           "exists": true,
-          "bytes": 7344,
-          "sha256": "d21fb26e1accaf0130f8ae0275526a49af52a6428b2fb0de7f8dc80d4efed230"
+          "bytes": 8930,
+          "sha256": "b4f0fd7644009ff4baeb98c05beb6b8bca86b166fdb6a7dc4f263a61a7ca13dd"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/project_status.json",
           "exists": true,
-          "bytes": 7344,
-          "sha256": "d21fb26e1accaf0130f8ae0275526a49af52a6428b2fb0de7f8dc80d4efed230"
+          "bytes": 8930,
+          "sha256": "b4f0fd7644009ff4baeb98c05beb6b8bca86b166fdb6a7dc4f263a61a7ca13dd"
         },
         "hf_model": {
           "path": "hf_model:metrics/project_status.json",
           "exists": true,
-          "bytes": 7344,
-          "sha256": "d21fb26e1accaf0130f8ae0275526a49af52a6428b2fb0de7f8dc80d4efed230"
+          "bytes": 8930,
+          "sha256": "b4f0fd7644009ff4baeb98c05beb6b8bca86b166fdb6a7dc4f263a61a7ca13dd"
         }
       },
       "failures": []
@@ -381,27 +443,27 @@
       "local": {
         "path": "repo:docs/data/publication_audit.json",
         "exists": true,
-        "bytes": 7253,
-        "sha256": "00f684ad0bf0a076817f37eb2962913b21e3520f06a1808dfaab725c2393ee78"
+        "bytes": 7289,
+        "sha256": "a019468bf2be313087ffa4b11088ccc633ac0482be38b2bf13e23540364be466"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/publication_audit.json",
           "exists": true,
-          "bytes": 7253,
-          "sha256": "00f684ad0bf0a076817f37eb2962913b21e3520f06a1808dfaab725c2393ee78"
+          "bytes": 7289,
+          "sha256": "a019468bf2be313087ffa4b11088ccc633ac0482be38b2bf13e23540364be466"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/publication_audit.json",
           "exists": true,
-          "bytes": 7253,
-          "sha256": "00f684ad0bf0a076817f37eb2962913b21e3520f06a1808dfaab725c2393ee78"
+          "bytes": 7289,
+          "sha256": "a019468bf2be313087ffa4b11088ccc633ac0482be38b2bf13e23540364be466"
         },
         "hf_model": {
           "path": "hf_model:metrics/publication_audit.json",
           "exists": true,
-          "bytes": 7253,
-          "sha256": "00f684ad0bf0a076817f37eb2962913b21e3520f06a1808dfaab725c2393ee78"
+          "bytes": 7289,
+          "sha256": "a019468bf2be313087ffa4b11088ccc633ac0482be38b2bf13e23540364be466"
         }
       },
       "failures": []
@@ -412,27 +474,27 @@
       "local": {
         "path": "repo:docs/data/public_surface_qa.json",
         "exists": true,
-        "bytes": 5651,
-        "sha256": "92ead5e68cf9a3da42c8095397f15be8a75d248b5371147f7806bc5d8e83e3ad"
+        "bytes": 5631,
+        "sha256": "0f61bee108f0dc72f48c4edd0ca194fc5d108bb0c5cb7bcc86c6eaa56f5fae64"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/public_surface_qa.json",
           "exists": true,
-          "bytes": 5651,
-          "sha256": "92ead5e68cf9a3da42c8095397f15be8a75d248b5371147f7806bc5d8e83e3ad"
+          "bytes": 5631,
+          "sha256": "0f61bee108f0dc72f48c4edd0ca194fc5d108bb0c5cb7bcc86c6eaa56f5fae64"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/public_surface_qa.json",
           "exists": true,
-          "bytes": 5651,
-          "sha256": "92ead5e68cf9a3da42c8095397f15be8a75d248b5371147f7806bc5d8e83e3ad"
+          "bytes": 5631,
+          "sha256": "0f61bee108f0dc72f48c4edd0ca194fc5d108bb0c5cb7bcc86c6eaa56f5fae64"
         },
         "hf_model": {
           "path": "hf_model:metrics/public_surface_qa.json",
           "exists": true,
-          "bytes": 5651,
-          "sha256": "92ead5e68cf9a3da42c8095397f15be8a75d248b5371147f7806bc5d8e83e3ad"
+          "bytes": 5631,
+          "sha256": "0f61bee108f0dc72f48c4edd0ca194fc5d108bb0c5cb7bcc86c6eaa56f5fae64"
         }
       },
       "failures": []
@@ -443,27 +505,27 @@
       "local": {
         "path": "repo:docs/data/quality_gates.json",
         "exists": true,
-        "bytes": 8147,
-        "sha256": "fddf86398d7187175c7321ea645c421ee599093fb1c484b1e8aa0f2a23174c3e"
+        "bytes": 8118,
+        "sha256": "595b2e050028b612b57ecfb5086c66bdf5040be90250430537cbd9e1b9c09e76"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/quality_gates.json",
           "exists": true,
-          "bytes": 8147,
-          "sha256": "fddf86398d7187175c7321ea645c421ee599093fb1c484b1e8aa0f2a23174c3e"
+          "bytes": 8118,
+          "sha256": "595b2e050028b612b57ecfb5086c66bdf5040be90250430537cbd9e1b9c09e76"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/quality_gates.json",
           "exists": true,
-          "bytes": 8147,
-          "sha256": "fddf86398d7187175c7321ea645c421ee599093fb1c484b1e8aa0f2a23174c3e"
+          "bytes": 8118,
+          "sha256": "595b2e050028b612b57ecfb5086c66bdf5040be90250430537cbd9e1b9c09e76"
         },
         "hf_model": {
           "path": "hf_model:metrics/quality_gates.json",
           "exists": true,
-          "bytes": 8147,
-          "sha256": "fddf86398d7187175c7321ea645c421ee599093fb1c484b1e8aa0f2a23174c3e"
+          "bytes": 8118,
+          "sha256": "595b2e050028b612b57ecfb5086c66bdf5040be90250430537cbd9e1b9c09e76"
         }
       },
       "failures": []
@@ -505,27 +567,27 @@
       "local": {
         "path": "repo:docs/data/reproducibility_matrix.json",
         "exists": true,
-        "bytes": 5195,
-        "sha256": "4ecf4f09138c58b3ebe6cfeab54907c372e717b2979df1341a6b3dfeb020b908"
+        "bytes": 5229,
+        "sha256": "928eb95b95d4028a7f21135584c37ff141833e26178cc918a874a6448f8c5ded"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5195,
-          "sha256": "4ecf4f09138c58b3ebe6cfeab54907c372e717b2979df1341a6b3dfeb020b908"
+          "bytes": 5229,
+          "sha256": "928eb95b95d4028a7f21135584c37ff141833e26178cc918a874a6448f8c5ded"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5195,
-          "sha256": "4ecf4f09138c58b3ebe6cfeab54907c372e717b2979df1341a6b3dfeb020b908"
+          "bytes": 5229,
+          "sha256": "928eb95b95d4028a7f21135584c37ff141833e26178cc918a874a6448f8c5ded"
         },
         "hf_model": {
           "path": "hf_model:metrics/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5195,
-          "sha256": "4ecf4f09138c58b3ebe6cfeab54907c372e717b2979df1341a6b3dfeb020b908"
+          "bytes": 5229,
+          "sha256": "928eb95b95d4028a7f21135584c37ff141833e26178cc918a874a6448f8c5ded"
         }
       },
       "failures": []
@@ -536,27 +598,27 @@
       "local": {
         "path": "repo:docs/data/research_roadmap.json",
         "exists": true,
-        "bytes": 4594,
-        "sha256": "bbce65e803a0fad55cc739eb2dc6fdf50a69c81362f05d216324d0813ee4ccad"
+        "bytes": 5726,
+        "sha256": "903ec0d6104c092fa0356c7741134fee4a1c78f3b0517001c49ce11f8077d6e2"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/research_roadmap.json",
           "exists": true,
-          "bytes": 4594,
-          "sha256": "bbce65e803a0fad55cc739eb2dc6fdf50a69c81362f05d216324d0813ee4ccad"
+          "bytes": 5726,
+          "sha256": "903ec0d6104c092fa0356c7741134fee4a1c78f3b0517001c49ce11f8077d6e2"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/research_roadmap.json",
           "exists": true,
-          "bytes": 4594,
-          "sha256": "bbce65e803a0fad55cc739eb2dc6fdf50a69c81362f05d216324d0813ee4ccad"
+          "bytes": 5726,
+          "sha256": "903ec0d6104c092fa0356c7741134fee4a1c78f3b0517001c49ce11f8077d6e2"
         },
         "hf_model": {
           "path": "hf_model:metrics/research_roadmap.json",
           "exists": true,
-          "bytes": 4594,
-          "sha256": "bbce65e803a0fad55cc739eb2dc6fdf50a69c81362f05d216324d0813ee4ccad"
+          "bytes": 5726,
+          "sha256": "903ec0d6104c092fa0356c7741134fee4a1c78f3b0517001c49ce11f8077d6e2"
         }
       },
       "failures": []
@@ -567,27 +629,27 @@
       "local": {
         "path": "repo:docs/data/research_roadmap_interactive.json",
         "exists": true,
-        "bytes": 73090,
-        "sha256": "204dbcaf022bb14f5d7ff7f68395788f4b6f4466cddffba73a7103d72ca43013"
+        "bytes": 131473,
+        "sha256": "c0e2f4872c7192995307967d27bf9c329ae88c685688e3f128f68e3fdc48e780"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/research_roadmap_interactive.json",
           "exists": true,
-          "bytes": 73090,
-          "sha256": "204dbcaf022bb14f5d7ff7f68395788f4b6f4466cddffba73a7103d72ca43013"
+          "bytes": 131473,
+          "sha256": "c0e2f4872c7192995307967d27bf9c329ae88c685688e3f128f68e3fdc48e780"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/research_roadmap_interactive.json",
           "exists": true,
-          "bytes": 73090,
-          "sha256": "204dbcaf022bb14f5d7ff7f68395788f4b6f4466cddffba73a7103d72ca43013"
+          "bytes": 131473,
+          "sha256": "c0e2f4872c7192995307967d27bf9c329ae88c685688e3f128f68e3fdc48e780"
         },
         "hf_model": {
           "path": "hf_model:metrics/research_roadmap_interactive.json",
           "exists": true,
-          "bytes": 73090,
-          "sha256": "204dbcaf022bb14f5d7ff7f68395788f4b6f4466cddffba73a7103d72ca43013"
+          "bytes": 131473,
+          "sha256": "c0e2f4872c7192995307967d27bf9c329ae88c685688e3f128f68e3fdc48e780"
         }
       },
       "failures": []
@@ -598,27 +660,27 @@
       "local": {
         "path": "repo:docs/data/research_takeaways.json",
         "exists": true,
-        "bytes": 5358,
-        "sha256": "f4c94ac0a7c162ff9f448635e54df125dbce47d2ad98fa8ca8d85eedb36e9959"
+        "bytes": 6821,
+        "sha256": "509d8690a4854097ea45af635dafa0231efa11838f1bb1f9cb338e99009a0206"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/research_takeaways.json",
           "exists": true,
-          "bytes": 5358,
-          "sha256": "f4c94ac0a7c162ff9f448635e54df125dbce47d2ad98fa8ca8d85eedb36e9959"
+          "bytes": 6821,
+          "sha256": "509d8690a4854097ea45af635dafa0231efa11838f1bb1f9cb338e99009a0206"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/research_takeaways.json",
           "exists": true,
-          "bytes": 5358,
-          "sha256": "f4c94ac0a7c162ff9f448635e54df125dbce47d2ad98fa8ca8d85eedb36e9959"
+          "bytes": 6821,
+          "sha256": "509d8690a4854097ea45af635dafa0231efa11838f1bb1f9cb338e99009a0206"
         },
         "hf_model": {
           "path": "hf_model:metrics/research_takeaways.json",
           "exists": true,
-          "bytes": 5358,
-          "sha256": "f4c94ac0a7c162ff9f448635e54df125dbce47d2ad98fa8ca8d85eedb36e9959"
+          "bytes": 6821,
+          "sha256": "509d8690a4854097ea45af635dafa0231efa11838f1bb1f9cb338e99009a0206"
         }
       },
       "failures": []
@@ -629,27 +691,27 @@
       "local": {
         "path": "repo:docs/data/research_direction_extensions.json",
         "exists": true,
-        "bytes": 11894,
-        "sha256": "ab0e51d53efe58a78f0f57aa683281e64701da86f6fa3ab1c185211b3c07164e"
+        "bytes": 11903,
+        "sha256": "010e254b61de6cc199e8e710fb10d8304c8156835f5cbafb79202996a74c0c77"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/research_direction_extensions.json",
           "exists": true,
-          "bytes": 11894,
-          "sha256": "ab0e51d53efe58a78f0f57aa683281e64701da86f6fa3ab1c185211b3c07164e"
+          "bytes": 11903,
+          "sha256": "010e254b61de6cc199e8e710fb10d8304c8156835f5cbafb79202996a74c0c77"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/research_direction_extensions.json",
           "exists": true,
-          "bytes": 11894,
-          "sha256": "ab0e51d53efe58a78f0f57aa683281e64701da86f6fa3ab1c185211b3c07164e"
+          "bytes": 11903,
+          "sha256": "010e254b61de6cc199e8e710fb10d8304c8156835f5cbafb79202996a74c0c77"
         },
         "hf_model": {
           "path": "hf_model:metrics/research_direction_extensions.json",
           "exists": true,
-          "bytes": 11894,
-          "sha256": "ab0e51d53efe58a78f0f57aa683281e64701da86f6fa3ab1c185211b3c07164e"
+          "bytes": 11903,
+          "sha256": "010e254b61de6cc199e8e710fb10d8304c8156835f5cbafb79202996a74c0c77"
         }
       },
       "failures": []
@@ -661,26 +723,26 @@
         "path": "repo:docs/data/research_directions.json",
         "exists": true,
         "bytes": 14414,
-        "sha256": "fbc45a85de7e52a1a67c413115bf062961bc00f413811f6d7951e4e1ac80824e"
+        "sha256": "7215681c55a6739da7f16d833f62fcb8f6d58069840963182e98427eaf0cf654"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/research_directions.json",
           "exists": true,
           "bytes": 14414,
-          "sha256": "fbc45a85de7e52a1a67c413115bf062961bc00f413811f6d7951e4e1ac80824e"
+          "sha256": "7215681c55a6739da7f16d833f62fcb8f6d58069840963182e98427eaf0cf654"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/research_directions.json",
           "exists": true,
           "bytes": 14414,
-          "sha256": "fbc45a85de7e52a1a67c413115bf062961bc00f413811f6d7951e4e1ac80824e"
+          "sha256": "7215681c55a6739da7f16d833f62fcb8f6d58069840963182e98427eaf0cf654"
         },
         "hf_model": {
           "path": "hf_model:metrics/research_directions.json",
           "exists": true,
           "bytes": 14414,
-          "sha256": "fbc45a85de7e52a1a67c413115bf062961bc00f413811f6d7951e4e1ac80824e"
+          "sha256": "7215681c55a6739da7f16d833f62fcb8f6d58069840963182e98427eaf0cf654"
         }
       },
       "failures": []
@@ -692,26 +754,26 @@
         "path": "repo:docs/data/scope_claims_audit.json",
         "exists": true,
         "bytes": 20066,
-        "sha256": "02d7c586e9e6a2d8ce99271bd0b13d6c1975dae6fa8c8ccafd398359e61d35c9"
+        "sha256": "93476525ab34432a84fc3093eee367b8dcbb913a30faa1244bfb6dba45d50161"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/scope_claims_audit.json",
           "exists": true,
           "bytes": 20066,
-          "sha256": "02d7c586e9e6a2d8ce99271bd0b13d6c1975dae6fa8c8ccafd398359e61d35c9"
+          "sha256": "93476525ab34432a84fc3093eee367b8dcbb913a30faa1244bfb6dba45d50161"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/scope_claims_audit.json",
           "exists": true,
           "bytes": 20066,
-          "sha256": "02d7c586e9e6a2d8ce99271bd0b13d6c1975dae6fa8c8ccafd398359e61d35c9"
+          "sha256": "93476525ab34432a84fc3093eee367b8dcbb913a30faa1244bfb6dba45d50161"
         },
         "hf_model": {
           "path": "hf_model:metrics/scope_claims_audit.json",
           "exists": true,
           "bytes": 20066,
-          "sha256": "02d7c586e9e6a2d8ce99271bd0b13d6c1975dae6fa8c8ccafd398359e61d35c9"
+          "sha256": "93476525ab34432a84fc3093eee367b8dcbb913a30faa1244bfb6dba45d50161"
         }
       },
       "failures": []
@@ -722,27 +784,27 @@
       "local": {
         "path": "repo:docs/data/single_episode_explorer.json",
         "exists": true,
-        "bytes": 4101241,
-        "sha256": "6940aa43d1b2d2b93c65a6e8abfbe5f252c8184372594483bc89341d1de6209d"
+        "bytes": 4297465,
+        "sha256": "89481cb9c11e923aef55dddd50b41a6fcc20ebf83efa94eb71df73b7782561f5"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/single_episode_explorer.json",
           "exists": true,
-          "bytes": 4101241,
-          "sha256": "6940aa43d1b2d2b93c65a6e8abfbe5f252c8184372594483bc89341d1de6209d"
+          "bytes": 4297465,
+          "sha256": "89481cb9c11e923aef55dddd50b41a6fcc20ebf83efa94eb71df73b7782561f5"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/single_episode_explorer.json",
           "exists": true,
-          "bytes": 4101241,
-          "sha256": "6940aa43d1b2d2b93c65a6e8abfbe5f252c8184372594483bc89341d1de6209d"
+          "bytes": 4297465,
+          "sha256": "89481cb9c11e923aef55dddd50b41a6fcc20ebf83efa94eb71df73b7782561f5"
         },
         "hf_model": {
           "path": "hf_model:metrics/single_episode_explorer.json",
           "exists": true,
-          "bytes": 4101241,
-          "sha256": "6940aa43d1b2d2b93c65a6e8abfbe5f252c8184372594483bc89341d1de6209d"
+          "bytes": 4297465,
+          "sha256": "89481cb9c11e923aef55dddd50b41a6fcc20ebf83efa94eb71df73b7782561f5"
         }
       },
       "failures": []
@@ -754,26 +816,26 @@
         "path": "repo:docs/data/source_alignment_audit.json",
         "exists": true,
         "bytes": 4432,
-        "sha256": "a9e89118de4da5899823cd9d5711dfd469b4a302e6483e059052ba3af8f76029"
+        "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "a9e89118de4da5899823cd9d5711dfd469b4a302e6483e059052ba3af8f76029"
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "a9e89118de4da5899823cd9d5711dfd469b4a302e6483e059052ba3af8f76029"
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
         },
         "hf_model": {
           "path": "hf_model:metrics/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "a9e89118de4da5899823cd9d5711dfd469b4a302e6483e059052ba3af8f76029"
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
         }
       },
       "failures": []
@@ -784,27 +846,27 @@
       "local": {
         "path": "repo:docs/data/summary_metrics.json",
         "exists": true,
-        "bytes": 25074,
-        "sha256": "3240ae9ca9109925bf0408a48d44817684a47a15ff942895220860b76284db1b"
+        "bytes": 25260,
+        "sha256": "e53e0a6151df2dd0bf02467421abf5a8e7655b88a21b6ce984786adca5d4cc43"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/summary_metrics.json",
           "exists": true,
-          "bytes": 25074,
-          "sha256": "3240ae9ca9109925bf0408a48d44817684a47a15ff942895220860b76284db1b"
+          "bytes": 25260,
+          "sha256": "e53e0a6151df2dd0bf02467421abf5a8e7655b88a21b6ce984786adca5d4cc43"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/summary_metrics.json",
           "exists": true,
-          "bytes": 25074,
-          "sha256": "3240ae9ca9109925bf0408a48d44817684a47a15ff942895220860b76284db1b"
+          "bytes": 25260,
+          "sha256": "e53e0a6151df2dd0bf02467421abf5a8e7655b88a21b6ce984786adca5d4cc43"
         },
         "hf_model": {
           "path": "hf_model:metrics/summary_metrics.json",
           "exists": true,
-          "bytes": 25074,
-          "sha256": "3240ae9ca9109925bf0408a48d44817684a47a15ff942895220860b76284db1b"
+          "bytes": 25260,
+          "sha256": "e53e0a6151df2dd0bf02467421abf5a8e7655b88a21b6ce984786adca5d4cc43"
         }
       },
       "failures": []
@@ -815,27 +877,27 @@
       "local": {
         "path": "repo:docs/data/task_surface_integrity.json",
         "exists": true,
-        "bytes": 45780,
-        "sha256": "f4d2894afa2aba013ed6728ee610d6665e08ae7baac9c28e94c86ac79144023e"
+        "bytes": 45779,
+        "sha256": "ea0a9de1e414e275b41eca27908c2d5575da3c05e05d7f2544122d763bbdfd33"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/task_surface_integrity.json",
           "exists": true,
-          "bytes": 45780,
-          "sha256": "f4d2894afa2aba013ed6728ee610d6665e08ae7baac9c28e94c86ac79144023e"
+          "bytes": 45779,
+          "sha256": "ea0a9de1e414e275b41eca27908c2d5575da3c05e05d7f2544122d763bbdfd33"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/task_surface_integrity.json",
           "exists": true,
-          "bytes": 45780,
-          "sha256": "f4d2894afa2aba013ed6728ee610d6665e08ae7baac9c28e94c86ac79144023e"
+          "bytes": 45779,
+          "sha256": "ea0a9de1e414e275b41eca27908c2d5575da3c05e05d7f2544122d763bbdfd33"
         },
         "hf_model": {
           "path": "hf_model:metrics/task_surface_integrity.json",
           "exists": true,
-          "bytes": 45780,
-          "sha256": "f4d2894afa2aba013ed6728ee610d6665e08ae7baac9c28e94c86ac79144023e"
+          "bytes": 45779,
+          "sha256": "ea0a9de1e414e275b41eca27908c2d5575da3c05e05d7f2544122d763bbdfd33"
         }
       },
       "failures": []
@@ -846,27 +908,27 @@
       "local": {
         "path": "repo:docs/data/task_walkthroughs.json",
         "exists": true,
-        "bytes": 26958,
-        "sha256": "2e9549fc3865070b554a4a43d022e500cc0aee5420c4406f4bef815554acaa52"
+        "bytes": 26963,
+        "sha256": "b93df37e7da1fcd5d3df7c2643b4659e2fef986dba52cb37a164874919ccc53a"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/task_walkthroughs.json",
           "exists": true,
-          "bytes": 26958,
-          "sha256": "2e9549fc3865070b554a4a43d022e500cc0aee5420c4406f4bef815554acaa52"
+          "bytes": 26963,
+          "sha256": "b93df37e7da1fcd5d3df7c2643b4659e2fef986dba52cb37a164874919ccc53a"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/task_walkthroughs.json",
           "exists": true,
-          "bytes": 26958,
-          "sha256": "2e9549fc3865070b554a4a43d022e500cc0aee5420c4406f4bef815554acaa52"
+          "bytes": 26963,
+          "sha256": "b93df37e7da1fcd5d3df7c2643b4659e2fef986dba52cb37a164874919ccc53a"
         },
         "hf_model": {
           "path": "hf_model:metrics/task_walkthroughs.json",
           "exists": true,
-          "bytes": 26958,
-          "sha256": "2e9549fc3865070b554a4a43d022e500cc0aee5420c4406f4bef815554acaa52"
+          "bytes": 26963,
+          "sha256": "b93df37e7da1fcd5d3df7c2643b4659e2fef986dba52cb37a164874919ccc53a"
         }
       },
       "failures": []
@@ -877,27 +939,27 @@
       "local": {
         "path": "repo:docs/data/website_integrity.json",
         "exists": true,
-        "bytes": 14310,
-        "sha256": "13e66632810e6c6f994f31f54e83b50ccbc790d277587f66c24221e9fa9d1ff9"
+        "bytes": 14718,
+        "sha256": "96372b1d3297dd81a393010264d4d92204bd140838aa7a81f227fea7d88b5fb9"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/website_integrity.json",
           "exists": true,
-          "bytes": 14310,
-          "sha256": "13e66632810e6c6f994f31f54e83b50ccbc790d277587f66c24221e9fa9d1ff9"
+          "bytes": 14718,
+          "sha256": "96372b1d3297dd81a393010264d4d92204bd140838aa7a81f227fea7d88b5fb9"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/website_integrity.json",
           "exists": true,
-          "bytes": 14310,
-          "sha256": "13e66632810e6c6f994f31f54e83b50ccbc790d277587f66c24221e9fa9d1ff9"
+          "bytes": 14718,
+          "sha256": "96372b1d3297dd81a393010264d4d92204bd140838aa7a81f227fea7d88b5fb9"
         },
         "hf_model": {
           "path": "hf_model:metrics/website_integrity.json",
           "exists": true,
-          "bytes": 14310,
-          "sha256": "13e66632810e6c6f994f31f54e83b50ccbc790d277587f66c24221e9fa9d1ff9"
+          "bytes": 14718,
+          "sha256": "96372b1d3297dd81a393010264d4d92204bd140838aa7a81f227fea7d88b5fb9"
         }
       },
       "failures": []
@@ -908,27 +970,64 @@
       "local": {
         "path": "repo:docs/data/xperience10m_dataset_card_alignment.json",
         "exists": true,
-        "bytes": 8037,
-        "sha256": "6e49a184495c1cc8061f68d7cc74b3ba6377aedffce90c203ae34168411fd538"
+        "bytes": 7559,
+        "sha256": "24ea3fc9e1ae8f698ca192ae1b8511dfe1129b807af237f41382e0644e5b58af"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/xperience10m_dataset_card_alignment.json",
           "exists": true,
-          "bytes": 8037,
-          "sha256": "6e49a184495c1cc8061f68d7cc74b3ba6377aedffce90c203ae34168411fd538"
+          "bytes": 7559,
+          "sha256": "24ea3fc9e1ae8f698ca192ae1b8511dfe1129b807af237f41382e0644e5b58af"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/xperience10m_dataset_card_alignment.json",
           "exists": true,
-          "bytes": 8037,
-          "sha256": "6e49a184495c1cc8061f68d7cc74b3ba6377aedffce90c203ae34168411fd538"
+          "bytes": 7559,
+          "sha256": "24ea3fc9e1ae8f698ca192ae1b8511dfe1129b807af237f41382e0644e5b58af"
         },
         "hf_model": {
           "path": "hf_model:metrics/xperience10m_dataset_card_alignment.json",
           "exists": true,
-          "bytes": 8037,
-          "sha256": "6e49a184495c1cc8061f68d7cc74b3ba6377aedffce90c203ae34168411fd538"
+          "bytes": 7559,
+          "sha256": "24ea3fc9e1ae8f698ca192ae1b8511dfe1129b807af237f41382e0644e5b58af"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "assets/charts/audio_ablation_delta.svg",
+      "status": "pass",
+      "local": {
+        "path": "repo:docs/assets/charts/audio_ablation_delta.svg",
+        "exists": true,
+        "bytes": 4146,
+        "sha256": "187dbabe01f9ff18841ff61a1e7fbf85bebdd188cc0f248bb5090d64528e7568"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:assets/charts/audio_ablation_delta.svg",
+          "exists": true,
+          "bytes": 4146,
+          "sha256": "187dbabe01f9ff18841ff61a1e7fbf85bebdd188cc0f248bb5090d64528e7568"
+        },
+        "hf_artifacts_docs": {
+          "path": "hf_artifacts:docs/assets/charts/audio_ablation_delta.svg",
+          "exists": true,
+          "bytes": 4146,
+          "sha256": "187dbabe01f9ff18841ff61a1e7fbf85bebdd188cc0f248bb5090d64528e7568"
+        },
+        "hf_artifacts_card": {
+          "path": "hf_artifacts:assets/charts/audio_ablation_delta.svg",
+          "exists": true,
+          "bytes": 4146,
+          "sha256": "187dbabe01f9ff18841ff61a1e7fbf85bebdd188cc0f248bb5090d64528e7568"
+        },
+        "hf_model": {
+          "path": "hf_model:assets/charts/audio_ablation_delta.svg",
+          "exists": true,
+          "bytes": 4146,
+          "sha256": "187dbabe01f9ff18841ff61a1e7fbf85bebdd188cc0f248bb5090d64528e7568"
         }
       },
       "failures": []
@@ -1198,33 +1297,33 @@
       "local": {
         "path": "repo:docs/assets/task_suite_infographic.png",
         "exists": true,
-        "bytes": 2619553,
-        "sha256": "20abc06b3e3976a323fb52e6cb0e4a0fbaf8759f1efcd98fb3b8ad9579dd3f00"
+        "bytes": 2612510,
+        "sha256": "213d81f49d27e3f2560c79e29a017c017cbe38d8d605815bf3bc87834a1424ae"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:assets/task_suite_infographic.png",
           "exists": true,
-          "bytes": 2619553,
-          "sha256": "20abc06b3e3976a323fb52e6cb0e4a0fbaf8759f1efcd98fb3b8ad9579dd3f00"
+          "bytes": 2612510,
+          "sha256": "213d81f49d27e3f2560c79e29a017c017cbe38d8d605815bf3bc87834a1424ae"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/assets/task_suite_infographic.png",
           "exists": true,
-          "bytes": 2619553,
-          "sha256": "20abc06b3e3976a323fb52e6cb0e4a0fbaf8759f1efcd98fb3b8ad9579dd3f00"
+          "bytes": 2612510,
+          "sha256": "213d81f49d27e3f2560c79e29a017c017cbe38d8d605815bf3bc87834a1424ae"
         },
         "hf_artifacts_card": {
           "path": "hf_artifacts:assets/task_suite_infographic.png",
           "exists": true,
-          "bytes": 2619553,
-          "sha256": "20abc06b3e3976a323fb52e6cb0e4a0fbaf8759f1efcd98fb3b8ad9579dd3f00"
+          "bytes": 2612510,
+          "sha256": "213d81f49d27e3f2560c79e29a017c017cbe38d8d605815bf3bc87834a1424ae"
         },
         "hf_model": {
           "path": "hf_model:assets/task_suite_infographic.png",
           "exists": true,
-          "bytes": 2619553,
-          "sha256": "20abc06b3e3976a323fb52e6cb0e4a0fbaf8759f1efcd98fb3b8ad9579dd3f00"
+          "bytes": 2612510,
+          "sha256": "213d81f49d27e3f2560c79e29a017c017cbe38d8d605815bf3bc87834a1424ae"
         }
       },
       "failures": []
@@ -1235,33 +1334,33 @@
       "local": {
         "path": "repo:docs/assets/pipeline_diagram.png",
         "exists": true,
-        "bytes": 705995,
-        "sha256": "e1429c6904941c7b2bd9e724b96681f518b07b6756c85ca5307f77b6b2818b32"
+        "bytes": 707296,
+        "sha256": "3eb7767a6ec78ef50847b79fad16e2438e84f449782f3fe8367417bece1e289c"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:assets/pipeline_diagram.png",
           "exists": true,
-          "bytes": 705995,
-          "sha256": "e1429c6904941c7b2bd9e724b96681f518b07b6756c85ca5307f77b6b2818b32"
+          "bytes": 707296,
+          "sha256": "3eb7767a6ec78ef50847b79fad16e2438e84f449782f3fe8367417bece1e289c"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/assets/pipeline_diagram.png",
           "exists": true,
-          "bytes": 705995,
-          "sha256": "e1429c6904941c7b2bd9e724b96681f518b07b6756c85ca5307f77b6b2818b32"
+          "bytes": 707296,
+          "sha256": "3eb7767a6ec78ef50847b79fad16e2438e84f449782f3fe8367417bece1e289c"
         },
         "hf_artifacts_card": {
           "path": "hf_artifacts:assets/pipeline_diagram.png",
           "exists": true,
-          "bytes": 705995,
-          "sha256": "e1429c6904941c7b2bd9e724b96681f518b07b6756c85ca5307f77b6b2818b32"
+          "bytes": 707296,
+          "sha256": "3eb7767a6ec78ef50847b79fad16e2438e84f449782f3fe8367417bece1e289c"
         },
         "hf_model": {
           "path": "hf_model:assets/pipeline_diagram.png",
           "exists": true,
-          "bytes": 705995,
-          "sha256": "e1429c6904941c7b2bd9e724b96681f518b07b6756c85ca5307f77b6b2818b32"
+          "bytes": 707296,
+          "sha256": "3eb7767a6ec78ef50847b79fad16e2438e84f449782f3fe8367417bece1e289c"
         }
       },
       "failures": []
@@ -1272,33 +1371,33 @@
       "local": {
         "path": "repo:docs/assets/task_architectures.png",
         "exists": true,
-        "bytes": 762376,
-        "sha256": "71bb1f6d7b9a65ec01f4cf9f79d35d22bb5932554d5dea42a8e8037db031e674"
+        "bytes": 767403,
+        "sha256": "9e3b86acf7f7296eb7c4b1e5e74c7008f264ddce40485e0e923b1e2a2f2b8efe"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:assets/task_architectures.png",
           "exists": true,
-          "bytes": 762376,
-          "sha256": "71bb1f6d7b9a65ec01f4cf9f79d35d22bb5932554d5dea42a8e8037db031e674"
+          "bytes": 767403,
+          "sha256": "9e3b86acf7f7296eb7c4b1e5e74c7008f264ddce40485e0e923b1e2a2f2b8efe"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/assets/task_architectures.png",
           "exists": true,
-          "bytes": 762376,
-          "sha256": "71bb1f6d7b9a65ec01f4cf9f79d35d22bb5932554d5dea42a8e8037db031e674"
+          "bytes": 767403,
+          "sha256": "9e3b86acf7f7296eb7c4b1e5e74c7008f264ddce40485e0e923b1e2a2f2b8efe"
         },
         "hf_artifacts_card": {
           "path": "hf_artifacts:assets/task_architectures.png",
           "exists": true,
-          "bytes": 762376,
-          "sha256": "71bb1f6d7b9a65ec01f4cf9f79d35d22bb5932554d5dea42a8e8037db031e674"
+          "bytes": 767403,
+          "sha256": "9e3b86acf7f7296eb7c4b1e5e74c7008f264ddce40485e0e923b1e2a2f2b8efe"
         },
         "hf_model": {
           "path": "hf_model:assets/task_architectures.png",
           "exists": true,
-          "bytes": 762376,
-          "sha256": "71bb1f6d7b9a65ec01f4cf9f79d35d22bb5932554d5dea42a8e8037db031e674"
+          "bytes": 767403,
+          "sha256": "9e3b86acf7f7296eb7c4b1e5e74c7008f264ddce40485e0e923b1e2a2f2b8efe"
         }
       },
       "failures": []
@@ -1562,27 +1661,52 @@
       },
       "failures": []
     },
+    {
+      "name": "scripts/audio_ablation_and_raw_upgrade.py",
+      "status": "pass",
+      "local": {
+        "path": "repo:scripts/audio_ablation_and_raw_upgrade.py",
+        "exists": true,
+        "bytes": 43144,
+        "sha256": "f7e3a38ec906dac7ca55b13c49720bd41ed89a1fd994c7d54730a4de5dfd1b59"
+      },
+      "mirrors": {
+        "hf_artifacts": {
+          "path": "hf_artifacts:scripts/audio_ablation_and_raw_upgrade.py",
+          "exists": true,
+          "bytes": 43144,
+          "sha256": "f7e3a38ec906dac7ca55b13c49720bd41ed89a1fd994c7d54730a4de5dfd1b59"
+        },
+        "hf_model": {
+          "path": "hf_model:scripts/audio_ablation_and_raw_upgrade.py",
+          "exists": true,
+          "bytes": 43144,
+          "sha256": "f7e3a38ec906dac7ca55b13c49720bd41ed89a1fd994c7d54730a4de5dfd1b59"
+        }
+      },
+      "failures": []
+    },
     {
       "name": "scripts/build_artifact_index.py",
       "status": "pass",
       "local": {
         "path": "repo:scripts/build_artifact_index.py",
         "exists": true,
-        "bytes": 23604,
-        "sha256": "e822eaa3fcbc805ac6869c16d46d41b78e1a1cfb921afa4956dc28c0a7c8daa3"
+        "bytes": 26566,
+        "sha256": "982fcc36e6dfda3382662d1f1e1f539f27f45aaddd6503aad77a14145bcc1bf6"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_artifact_index.py",
           "exists": true,
-          "bytes": 23604,
-          "sha256": "e822eaa3fcbc805ac6869c16d46d41b78e1a1cfb921afa4956dc28c0a7c8daa3"
+          "bytes": 26566,
+          "sha256": "982fcc36e6dfda3382662d1f1e1f539f27f45aaddd6503aad77a14145bcc1bf6"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_artifact_index.py",
           "exists": true,
-          "bytes": 23604,
-          "sha256": "e822eaa3fcbc805ac6869c16d46d41b78e1a1cfb921afa4956dc28c0a7c8daa3"
+          "bytes": 26566,
+          "sha256": "982fcc36e6dfda3382662d1f1e1f539f27f45aaddd6503aad77a14145bcc1bf6"
         }
       },
       "failures": []
@@ -1618,21 +1742,21 @@
       "local": {
         "path": "repo:scripts/build_evaluation_protocol.py",
         "exists": true,
-        "bytes": 16168,
-        "sha256": "01bf9dd4de2970636ec558104cd908beab6b2f8dd6290c60f5f4c2693dd55a94"
+        "bytes": 16139,
+        "sha256": "e2aaac38f451402971e03c004d32d5c5f8a05d67dc1820d6bf9b12d1da1a292f"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_evaluation_protocol.py",
           "exists": true,
-          "bytes": 16168,
-          "sha256": "01bf9dd4de2970636ec558104cd908beab6b2f8dd6290c60f5f4c2693dd55a94"
+          "bytes": 16139,
+          "sha256": "e2aaac38f451402971e03c004d32d5c5f8a05d67dc1820d6bf9b12d1da1a292f"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_evaluation_protocol.py",
           "exists": true,
-          "bytes": 16168,
-          "sha256": "01bf9dd4de2970636ec558104cd908beab6b2f8dd6290c60f5f4c2693dd55a94"
+          "bytes": 16139,
+          "sha256": "e2aaac38f451402971e03c004d32d5c5f8a05d67dc1820d6bf9b12d1da1a292f"
         }
       },
       "failures": []
@@ -1668,21 +1792,21 @@
       "local": {
         "path": "repo:scripts/build_quality_gates.py",
         "exists": true,
-        "bytes": 11380,
-        "sha256": "e8c99db9b2698f824b580882b238a881cffeafffc0b5f768d5fad0f9ac24ae18"
+        "bytes": 11351,
+        "sha256": "1a185999e568243fb480ba21d1fc181037a97136dec4e562b3b78fb1dd7d9ead"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_quality_gates.py",
           "exists": true,
-          "bytes": 11380,
-          "sha256": "e8c99db9b2698f824b580882b238a881cffeafffc0b5f768d5fad0f9ac24ae18"
+          "bytes": 11351,
+          "sha256": "1a185999e568243fb480ba21d1fc181037a97136dec4e562b3b78fb1dd7d9ead"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_quality_gates.py",
           "exists": true,
-          "bytes": 11380,
-          "sha256": "e8c99db9b2698f824b580882b238a881cffeafffc0b5f768d5fad0f9ac24ae18"
+          "bytes": 11351,
+          "sha256": "1a185999e568243fb480ba21d1fc181037a97136dec4e562b3b78fb1dd7d9ead"
         }
       },
       "failures": []
@@ -1693,21 +1817,21 @@
       "local": {
         "path": "repo:scripts/build_public_surface_qa.py",
         "exists": true,
-        "bytes": 11894,
-        "sha256": "d0e86f45f2f23670e967c1a92024797a59106116ac9d948c35972512384f41de"
+        "bytes": 11872,
+        "sha256": "3ec688d5262d099b480bda674b858834d5d66b8085c0777a2435e58279f37d9d"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_public_surface_qa.py",
           "exists": true,
-          "bytes": 11894,
-          "sha256": "d0e86f45f2f23670e967c1a92024797a59106116ac9d948c35972512384f41de"
+          "bytes": 11872,
+          "sha256": "3ec688d5262d099b480bda674b858834d5d66b8085c0777a2435e58279f37d9d"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_public_surface_qa.py",
           "exists": true,
-          "bytes": 11894,
-          "sha256": "d0e86f45f2f23670e967c1a92024797a59106116ac9d948c35972512384f41de"
+          "bytes": 11872,
+          "sha256": "3ec688d5262d099b480bda674b858834d5d66b8085c0777a2435e58279f37d9d"
         }
       },
       "failures": []
@@ -1743,21 +1867,21 @@
       "local": {
         "path": "repo:scripts/build_interactive_research_roadmap.py",
         "exists": true,
-        "bytes": 8339,
-        "sha256": "3718081b1446086576d683f9b76148857b34c5d0022a101df950f7cb5679d5da"
+        "bytes": 10167,
+        "sha256": "36d860fe9434937e0b6b8b342013409344c27f6cae5dbc5758b4e58d104ba799"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_interactive_research_roadmap.py",
           "exists": true,
-          "bytes": 8339,
-          "sha256": "3718081b1446086576d683f9b76148857b34c5d0022a101df950f7cb5679d5da"
+          "bytes": 10167,
+          "sha256": "36d860fe9434937e0b6b8b342013409344c27f6cae5dbc5758b4e58d104ba799"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_interactive_research_roadmap.py",
           "exists": true,
-          "bytes": 8339,
-          "sha256": "3718081b1446086576d683f9b76148857b34c5d0022a101df950f7cb5679d5da"
+          "bytes": 10167,
+          "sha256": "36d860fe9434937e0b6b8b342013409344c27f6cae5dbc5758b4e58d104ba799"
         }
       },
       "failures": []
@@ -1768,21 +1892,21 @@
       "local": {
         "path": "repo:scripts/build_single_episode_explorer.py",
         "exists": true,
-        "bytes": 30104,
-        "sha256": "3412b7e0396a5471feae75054f6262993bfd3529096cd919cb03cc2edfd3b253"
+        "bytes": 30200,
+        "sha256": "57924e42243c76fb37dc36acb1c41932edd0467da7b2794ea6bd08603884b61a"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_single_episode_explorer.py",
           "exists": true,
-          "bytes": 30104,
-          "sha256": "3412b7e0396a5471feae75054f6262993bfd3529096cd919cb03cc2edfd3b253"
+          "bytes": 30200,
+          "sha256": "57924e42243c76fb37dc36acb1c41932edd0467da7b2794ea6bd08603884b61a"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_single_episode_explorer.py",
           "exists": true,
-          "bytes": 30104,
-          "sha256": "3412b7e0396a5471feae75054f6262993bfd3529096cd919cb03cc2edfd3b253"
+          "bytes": 30200,
+          "sha256": "57924e42243c76fb37dc36acb1c41932edd0467da7b2794ea6bd08603884b61a"
         }
       },
       "failures": []
@@ -1793,21 +1917,21 @@
       "local": {
         "path": "repo:scripts/build_research_takeaways.py",
         "exists": true,
-        "bytes": 10667,
-        "sha256": "a14855d74b6c6fc8291821b9fbba9185623533ff2e21c96ec53d84071c9e4eae"
+        "bytes": 13285,
+        "sha256": "19baa1dbe3e4454e040bb1b2fc711e1dd5ce5280ee19bfc01b01cca854006d38"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_research_takeaways.py",
           "exists": true,
-          "bytes": 10667,
-          "sha256": "a14855d74b6c6fc8291821b9fbba9185623533ff2e21c96ec53d84071c9e4eae"
+          "bytes": 13285,
+          "sha256": "19baa1dbe3e4454e040bb1b2fc711e1dd5ce5280ee19bfc01b01cca854006d38"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_research_takeaways.py",
           "exists": true,
-          "bytes": 10667,
-          "sha256": "a14855d74b6c6fc8291821b9fbba9185623533ff2e21c96ec53d84071c9e4eae"
+          "bytes": 13285,
+          "sha256": "19baa1dbe3e4454e040bb1b2fc711e1dd5ce5280ee19bfc01b01cca854006d38"
         }
       },
       "failures": []
@@ -1818,21 +1942,21 @@
       "local": {
         "path": "repo:scripts/single_episode_diagnostics.py",
         "exists": true,
-        "bytes": 57663,
-        "sha256": "6387a8da67d9a6d8c5bebaa25efc66371f8f64d891a949e23ad9cfeccb202770"
+        "bytes": 57814,
+        "sha256": "51523e67db90c1dd2b803d6e4f81ba2dbe5d059c1ffb95ede4f74cf4d900bec3"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/single_episode_diagnostics.py",
           "exists": true,
-          "bytes": 57663,
-          "sha256": "6387a8da67d9a6d8c5bebaa25efc66371f8f64d891a949e23ad9cfeccb202770"
+          "bytes": 57814,
+          "sha256": "51523e67db90c1dd2b803d6e4f81ba2dbe5d059c1ffb95ede4f74cf4d900bec3"
         },
         "hf_model": {
           "path": "hf_model:scripts/single_episode_diagnostics.py",
           "exists": true,
-          "bytes": 57663,
-          "sha256": "6387a8da67d9a6d8c5bebaa25efc66371f8f64d891a949e23ad9cfeccb202770"
+          "bytes": 57814,
+          "sha256": "51523e67db90c1dd2b803d6e4f81ba2dbe5d059c1ffb95ede4f74cf4d900bec3"
         }
       },
       "failures": []
@@ -1843,21 +1967,21 @@
       "local": {
         "path": "repo:scripts/verify_live_publication.py",
         "exists": true,
-        "bytes": 34767,
-        "sha256": "6a1cb0531719888de6bbeb7a2628058f6e67c92df6314c4ac3952af2c17f126b"
+        "bytes": 34779,
+        "sha256": "73c45e1ae1b7509b9f2c54cf42424c9ac0aab8a9e54be58fc488eaa2a696acc7"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/verify_live_publication.py",
           "exists": true,
-          "bytes": 34767,
-          "sha256": "6a1cb0531719888de6bbeb7a2628058f6e67c92df6314c4ac3952af2c17f126b"
+          "bytes": 34779,
+          "sha256": "73c45e1ae1b7509b9f2c54cf42424c9ac0aab8a9e54be58fc488eaa2a696acc7"
         },
         "hf_model": {
           "path": "hf_model:scripts/verify_live_publication.py",
           "exists": true,
-          "bytes": 34767,
-          "sha256": "6a1cb0531719888de6bbeb7a2628058f6e67c92df6314c4ac3952af2c17f126b"
+          "bytes": 34779,
+          "sha256": "73c45e1ae1b7509b9f2c54cf42424c9ac0aab8a9e54be58fc488eaa2a696acc7"
         }
       },
       "failures": []
@@ -1868,21 +1992,21 @@
       "local": {
         "path": "repo:scripts/validate_mirror_parity.py",
         "exists": true,
-        "bytes": 12057,
-        "sha256": "e2d87770613a36c618bc16c4d72e08064c397eef5860594f04f501446d54ed65"
+        "bytes": 12510,
+        "sha256": "4117c8e845d2cb09994438dd1c57d7ed5f2f69e38876506609271f83463bea25"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/validate_mirror_parity.py",
           "exists": true,
-          "bytes": 12057,
-          "sha256": "e2d87770613a36c618bc16c4d72e08064c397eef5860594f04f501446d54ed65"
+          "bytes": 12510,
+          "sha256": "4117c8e845d2cb09994438dd1c57d7ed5f2f69e38876506609271f83463bea25"
         },
         "hf_model": {
           "path": "hf_model:scripts/validate_mirror_parity.py",
           "exists": true,
-          "bytes": 12057,
-          "sha256": "e2d87770613a36c618bc16c4d72e08064c397eef5860594f04f501446d54ed65"
+          "bytes": 12510,
+          "sha256": "4117c8e845d2cb09994438dd1c57d7ed5f2f69e38876506609271f83463bea25"
         }
       },
       "failures": []
@@ -1893,21 +2017,21 @@
       "local": {
         "path": "repo:scripts/validate_publication_package.py",
         "exists": true,
-        "bytes": 19247,
-        "sha256": "55a9c51d57766d54f9eb884e06d644fd9f3d6e358b80c06e7f515ab764cc2158"
+        "bytes": 19276,
+        "sha256": "861b599ed7df030942e179cb8e9e68b5395a11e97d01d2775ecd5c9c2a3916d2"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/validate_publication_package.py",
           "exists": true,
-          "bytes": 19247,
-          "sha256": "55a9c51d57766d54f9eb884e06d644fd9f3d6e358b80c06e7f515ab764cc2158"
+          "bytes": 19276,
+          "sha256": "861b599ed7df030942e179cb8e9e68b5395a11e97d01d2775ecd5c9c2a3916d2"
         },
         "hf_model": {
           "path": "hf_model:scripts/validate_publication_package.py",
           "exists": true,
-          "bytes": 19247,
-          "sha256": "55a9c51d57766d54f9eb884e06d644fd9f3d6e358b80c06e7f515ab764cc2158"
+          "bytes": 19276,
+          "sha256": "861b599ed7df030942e179cb8e9e68b5395a11e97d01d2775ecd5c9c2a3916d2"
         }
       },
       "failures": []
@@ -1943,21 +2067,21 @@
       "local": {
         "path": "repo:scripts/validate_source_alignment.py",
         "exists": true,
-        "bytes": 15801,
-        "sha256": "cb2d11aba5dcadf030c564288d6b502ab3e46bed2486ecd7c6d7b200d22ad773"
+        "bytes": 15831,
+        "sha256": "6f48190ca77dd6ba9dd2f821d5ace7e33bd9b4d9d76adb7237cc51ddb88de239"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/validate_source_alignment.py",
           "exists": true,
-          "bytes": 15801,
-          "sha256": "cb2d11aba5dcadf030c564288d6b502ab3e46bed2486ecd7c6d7b200d22ad773"
+          "bytes": 15831,
+          "sha256": "6f48190ca77dd6ba9dd2f821d5ace7e33bd9b4d9d76adb7237cc51ddb88de239"
         },
         "hf_model": {
           "path": "hf_model:scripts/validate_source_alignment.py",
           "exists": true,
-          "bytes": 15801,
-          "sha256": "cb2d11aba5dcadf030c564288d6b502ab3e46bed2486ecd7c6d7b200d22ad773"
+          "bytes": 15831,
+          "sha256": "6f48190ca77dd6ba9dd2f821d5ace7e33bd9b4d9d76adb7237cc51ddb88de239"
         }
       },
       "failures": []
@@ -2018,21 +2142,21 @@
       "local": {
         "path": "repo:scripts/publish_hf_bundles.py",
         "exists": true,
-        "bytes": 8996,
-        "sha256": "9790951493763e6490d01fec62661df2bbe2a3341466fe41298b95cda0900229"
+        "bytes": 9997,
+        "sha256": "b1482905fac18e598337b2f42028f7ffa46ba71a28726e473dc7cd2af1c70f9c"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/publish_hf_bundles.py",
           "exists": true,
-          "bytes": 8996,
-          "sha256": "9790951493763e6490d01fec62661df2bbe2a3341466fe41298b95cda0900229"
+          "bytes": 9997,
+          "sha256": "b1482905fac18e598337b2f42028f7ffa46ba71a28726e473dc7cd2af1c70f9c"
         },
         "hf_model": {
           "path": "hf_model:scripts/publish_hf_bundles.py",
           "exists": true,
-          "bytes": 8996,
-          "sha256": "9790951493763e6490d01fec62661df2bbe2a3341466fe41298b95cda0900229"
+          "bytes": 9997,
+          "sha256": "b1482905fac18e598337b2f42028f7ffa46ba71a28726e473dc7cd2af1c70f9c"
         }
       },
       "failures": []
@@ -2093,21 +2217,21 @@
       "local": {
         "path": "repo:docs/index.html",
         "exists": true,
-        "bytes": 166126,
-        "sha256": "2a53cbfacf2330f6b345d213918a892841719e03c9437198907e2f7fb4da6220"
+        "bytes": 170233,
+        "sha256": "669d8f0e4f278802df86787e4e69efa87e301091bb4c83ac388670c0d456c881"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:index.html",
           "exists": true,
-          "bytes": 166126,
-          "sha256": "2a53cbfacf2330f6b345d213918a892841719e03c9437198907e2f7fb4da6220"
+          "bytes": 170233,
+          "sha256": "669d8f0e4f278802df86787e4e69efa87e301091bb4c83ac388670c0d456c881"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/index.html",
           "exists": true,
-          "bytes": 166126,
-          "sha256": "2a53cbfacf2330f6b345d213918a892841719e03c9437198907e2f7fb4da6220"
+          "bytes": 170233,
+          "sha256": "669d8f0e4f278802df86787e4e69efa87e301091bb4c83ac388670c0d456c881"
         }
       },
       "failures": []
@@ -2118,21 +2242,21 @@
       "local": {
         "path": "repo:docs/research_roadmap.html",
         "exists": true,
-        "bytes": 31119,
-        "sha256": "c190599a74cf40f16d6d554baddbc9f47fabb7e022280399ef06c0d3de1b2047"
+        "bytes": 31593,
+        "sha256": "05864ee10e582a4c05a5f1cdda4ae92a387c457375f4c7631fd303de96ad0c98"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:research_roadmap.html",
           "exists": true,
-          "bytes": 31119,
-          "sha256": "c190599a74cf40f16d6d554baddbc9f47fabb7e022280399ef06c0d3de1b2047"
+          "bytes": 31593,
+          "sha256": "05864ee10e582a4c05a5f1cdda4ae92a387c457375f4c7631fd303de96ad0c98"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/research_roadmap.html",
           "exists": true,
-          "bytes": 31119,
-          "sha256": "c190599a74cf40f16d6d554baddbc9f47fabb7e022280399ef06c0d3de1b2047"
+          "bytes": 31593,
+          "sha256": "05864ee10e582a4c05a5f1cdda4ae92a387c457375f4c7631fd303de96ad0c98"
         }
       },
       "failures": []
@@ -2143,21 +2267,21 @@
       "local": {
         "path": "repo:docs/single_episode_explorer.html",
         "exists": true,
-        "bytes": 2642212,
-        "sha256": "7301e2b483fd2ff516952c74e2f37399cab3245a7fb1ec72f3525fa0daa520dd"
+        "bytes": 2766828,
+        "sha256": "19feb5e101ef50ec9f2a03e608ac75da68d642a2a38f20bd475635f0928279f0"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:single_episode_explorer.html",
           "exists": true,
-          "bytes": 2642212,
-          "sha256": "7301e2b483fd2ff516952c74e2f37399cab3245a7fb1ec72f3525fa0daa520dd"
+          "bytes": 2766828,
+          "sha256": "19feb5e101ef50ec9f2a03e608ac75da68d642a2a38f20bd475635f0928279f0"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/single_episode_explorer.html",
           "exists": true,
-          "bytes": 2642212,
-          "sha256": "7301e2b483fd2ff516952c74e2f37399cab3245a7fb1ec72f3525fa0daa520dd"
+          "bytes": 2766828,
+          "sha256": "19feb5e101ef50ec9f2a03e608ac75da68d642a2a38f20bd475635f0928279f0"
         }
       },
       "failures": []
@@ -2187,33 +2311,188 @@
       },
       "failures": []
     },
+    {
+      "name": "results/audio_ablation/AUDIO_ABLATION_SUMMARY.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/audio_ablation/AUDIO_ABLATION_SUMMARY.md",
+        "exists": true,
+        "bytes": 2166,
+        "sha256": "9318bc55b1b051c384ce2fd6e7cd0a56becfa00185a0b597fdc474a40b1f17f9"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/audio_ablation/AUDIO_ABLATION_SUMMARY.md",
+          "exists": true,
+          "bytes": 2166,
+          "sha256": "9318bc55b1b051c384ce2fd6e7cd0a56becfa00185a0b597fdc474a40b1f17f9"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/audio_ablation/AUDIO_ABLATION_SUMMARY.md",
+          "exists": true,
+          "bytes": 2166,
+          "sha256": "9318bc55b1b051c384ce2fd6e7cd0a56becfa00185a0b597fdc474a40b1f17f9"
+        },
+        "hf_model": {
+          "path": "hf_model:results/audio_ablation/AUDIO_ABLATION_SUMMARY.md",
+          "exists": true,
+          "bytes": 2166,
+          "sha256": "9318bc55b1b051c384ce2fd6e7cd0a56becfa00185a0b597fdc474a40b1f17f9"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/audio_ablation/audio_ablation_metrics.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/audio_ablation/audio_ablation_metrics.csv",
+        "exists": true,
+        "bytes": 24446,
+        "sha256": "674efea54e1be2c835d6c473c9df54ef02fc34e5f2be37728722d914892c8854"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/audio_ablation/audio_ablation_metrics.csv",
+          "exists": true,
+          "bytes": 24446,
+          "sha256": "674efea54e1be2c835d6c473c9df54ef02fc34e5f2be37728722d914892c8854"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/audio_ablation/audio_ablation_metrics.csv",
+          "exists": true,
+          "bytes": 24446,
+          "sha256": "674efea54e1be2c835d6c473c9df54ef02fc34e5f2be37728722d914892c8854"
+        },
+        "hf_model": {
+          "path": "hf_model:results/audio_ablation/audio_ablation_metrics.csv",
+          "exists": true,
+          "bytes": 24446,
+          "sha256": "674efea54e1be2c835d6c473c9df54ef02fc34e5f2be37728722d914892c8854"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/audio_ablation/audio_ablation_summary.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/audio_ablation/audio_ablation_summary.json",
+        "exists": true,
+        "bytes": 9735,
+        "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/audio_ablation/audio_ablation_summary.json",
+          "exists": true,
+          "bytes": 9735,
+          "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/audio_ablation/audio_ablation_summary.json",
+          "exists": true,
+          "bytes": 9735,
+          "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
+        },
+        "hf_model": {
+          "path": "hf_model:results/audio_ablation/audio_ablation_summary.json",
+          "exists": true,
+          "bytes": 9735,
+          "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/audio_ablation/audio_delta_summary.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/audio_ablation/audio_delta_summary.csv",
+        "exists": true,
+        "bytes": 2630,
+        "sha256": "55638ba2387085e3d21fd58be05de060da572ae009b8019926ab4e1733540f91"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/audio_ablation/audio_delta_summary.csv",
+          "exists": true,
+          "bytes": 2630,
+          "sha256": "55638ba2387085e3d21fd58be05de060da572ae009b8019926ab4e1733540f91"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/audio_ablation/audio_delta_summary.csv",
+          "exists": true,
+          "bytes": 2630,
+          "sha256": "55638ba2387085e3d21fd58be05de060da572ae009b8019926ab4e1733540f91"
+        },
+        "hf_model": {
+          "path": "hf_model:results/audio_ablation/audio_delta_summary.csv",
+          "exists": true,
+          "bytes": 2630,
+          "sha256": "55638ba2387085e3d21fd58be05de060da572ae009b8019926ab4e1733540f91"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/audio_ablation/raw_logmel_fisheye_cam0_sr16000_mels64_fft512_hop160.npz",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/audio_ablation/raw_logmel_fisheye_cam0_sr16000_mels64_fft512_hop160.npz",
+        "exists": true,
+        "bytes": 2363664,
+        "sha256": "6b61a3db5d21c89ba8cfb66b94006cb48640c5ba54d7d6ccd8835d7942534fa1"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/audio_ablation/raw_logmel_fisheye_cam0_sr16000_mels64_fft512_hop160.npz",
+          "exists": true,
+          "bytes": 2363664,
+          "sha256": "6b61a3db5d21c89ba8cfb66b94006cb48640c5ba54d7d6ccd8835d7942534fa1"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/audio_ablation/raw_logmel_fisheye_cam0_sr16000_mels64_fft512_hop160.npz",
+          "exists": true,
+          "bytes": 2363664,
+          "sha256": "6b61a3db5d21c89ba8cfb66b94006cb48640c5ba54d7d6ccd8835d7942534fa1"
+        },
+        "hf_model": {
+          "path": "hf_model:results/audio_ablation/raw_logmel_fisheye_cam0_sr16000_mels64_fft512_hop160.npz",
+          "exists": true,
+          "bytes": 2363664,
+          "sha256": "6b61a3db5d21c89ba8cfb66b94006cb48640c5ba54d7d6ccd8835d7942534fa1"
+        }
+      },
+      "failures": []
+    },
     {
       "name": "results/single_episode_diagnostics/provenance.json",
       "status": "pass",
       "local": {
         "path": "repo:results/single_episode_diagnostics/provenance.json",
         "exists": true,
-        "bytes": 3962,
-        "sha256": "48087224240941fd92444d4fee94a6146c96c35b5acc429a209db3c5cdb8d24b"
+        "bytes": 4072,
+        "sha256": "7189be7fc773943859dc75a6119c1fb654d8905ea381e8fc9f782d4b33f5e466"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/single_episode_diagnostics/provenance.json",
           "exists": true,
-          "bytes": 3962,
-          "sha256": "48087224240941fd92444d4fee94a6146c96c35b5acc429a209db3c5cdb8d24b"
+          "bytes": 4072,
+          "sha256": "7189be7fc773943859dc75a6119c1fb654d8905ea381e8fc9f782d4b33f5e466"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:results/single_episode_diagnostics/provenance.json",
           "exists": true,
-          "bytes": 3962,
-          "sha256": "48087224240941fd92444d4fee94a6146c96c35b5acc429a209db3c5cdb8d24b"
+          "bytes": 4072,
+          "sha256": "7189be7fc773943859dc75a6119c1fb654d8905ea381e8fc9f782d4b33f5e466"
         },
         "hf_model": {
           "path": "hf_model:results/single_episode_diagnostics/provenance.json",
           "exists": true,
-          "bytes": 3962,
-          "sha256": "48087224240941fd92444d4fee94a6146c96c35b5acc429a209db3c5cdb8d24b"
+          "bytes": 4072,
+          "sha256": "7189be7fc773943859dc75a6119c1fb654d8905ea381e8fc9f782d4b33f5e466"
         }
       },
       "failures": []
@@ -2224,27 +2503,27 @@
       "local": {
         "path": "repo:results/single_episode_diagnostics/README.md",
         "exists": true,
-        "bytes": 1058,
-        "sha256": "e64b13f227f270dc545317636f432fb766d6f7ab8695f69234b2c24b6413a42e"
+        "bytes": 1060,
+        "sha256": "95b9246a2fdeeb0ab234057a6238cd01d70e208f8fdcb340089e8682aac6c9f9"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/single_episode_diagnostics/README.md",
           "exists": true,
-          "bytes": 1058,
-          "sha256": "e64b13f227f270dc545317636f432fb766d6f7ab8695f69234b2c24b6413a42e"
+          "bytes": 1060,
+          "sha256": "95b9246a2fdeeb0ab234057a6238cd01d70e208f8fdcb340089e8682aac6c9f9"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:results/single_episode_diagnostics/README.md",
           "exists": true,
-          "bytes": 1058,
-          "sha256": "e64b13f227f270dc545317636f432fb766d6f7ab8695f69234b2c24b6413a42e"
+          "bytes": 1060,
+          "sha256": "95b9246a2fdeeb0ab234057a6238cd01d70e208f8fdcb340089e8682aac6c9f9"
         },
         "hf_model": {
           "path": "hf_model:results/single_episode_diagnostics/README.md",
           "exists": true,
-          "bytes": 1058,
-          "sha256": "e64b13f227f270dc545317636f432fb766d6f7ab8695f69234b2c24b6413a42e"
+          "bytes": 1060,
+          "sha256": "95b9246a2fdeeb0ab234057a6238cd01d70e208f8fdcb340089e8682aac6c9f9"
         }
       },
       "failures": []
@@ -2255,27 +2534,27 @@
       "local": {
         "path": "repo:results/single_episode_diagnostics/modality_ablation/ablation_metrics.csv",
         "exists": true,
-        "bytes": 24871,
-        "sha256": "8e43bfa9b90544d7b09e207de26d1b852dd115b3e7c957eaa4b45e520dc050b8"
+        "bytes": 27893,
+        "sha256": "8ea11ccb28a6d622d4b729fe7ff447e315a50784631d69cdb06f39857284a780"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/single_episode_diagnostics/modality_ablation/ablation_metrics.csv",
           "exists": true,
-          "bytes": 24871,
-          "sha256": "8e43bfa9b90544d7b09e207de26d1b852dd115b3e7c957eaa4b45e520dc050b8"
+          "bytes": 27893,
+          "sha256": "8ea11ccb28a6d622d4b729fe7ff447e315a50784631d69cdb06f39857284a780"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:results/single_episode_diagnostics/modality_ablation/ablation_metrics.csv",
           "exists": true,
-          "bytes": 24871,
-          "sha256": "8e43bfa9b90544d7b09e207de26d1b852dd115b3e7c957eaa4b45e520dc050b8"
+          "bytes": 27893,
+          "sha256": "8ea11ccb28a6d622d4b729fe7ff447e315a50784631d69cdb06f39857284a780"
         },
         "hf_model": {
           "path": "hf_model:results/single_episode_diagnostics/modality_ablation/ablation_metrics.csv",
           "exists": true,
-          "bytes": 24871,
-          "sha256": "8e43bfa9b90544d7b09e207de26d1b852dd115b3e7c957eaa4b45e520dc050b8"
+          "bytes": 27893,
+          "sha256": "8ea11ccb28a6d622d4b729fe7ff447e315a50784631d69cdb06f39857284a780"
         }
       },
       "failures": []
@@ -2286,27 +2565,27 @@
       "local": {
         "path": "repo:results/single_episode_diagnostics/modality_ablation/ablation_summary.json",
         "exists": true,
-        "bytes": 735,
-        "sha256": "2761e1777963473ffa5110ab25e863a6b8ee19ce2004d94ffa938cfa5e0d93fc"
+        "bytes": 755,
+        "sha256": "412b740e36b90f7fbad15c4c4fc46832ab8119d0c84f734b975a5a350836799f"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/single_episode_diagnostics/modality_ablation/ablation_summary.json",
           "exists": true,
-          "bytes": 735,
-          "sha256": "2761e1777963473ffa5110ab25e863a6b8ee19ce2004d94ffa938cfa5e0d93fc"
+          "bytes": 755,
+          "sha256": "412b740e36b90f7fbad15c4c4fc46832ab8119d0c84f734b975a5a350836799f"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:results/single_episode_diagnostics/modality_ablation/ablation_summary.json",
           "exists": true,
-          "bytes": 735,
-          "sha256": "2761e1777963473ffa5110ab25e863a6b8ee19ce2004d94ffa938cfa5e0d93fc"
+          "bytes": 755,
+          "sha256": "412b740e36b90f7fbad15c4c4fc46832ab8119d0c84f734b975a5a350836799f"
         },
         "hf_model": {
           "path": "hf_model:results/single_episode_diagnostics/modality_ablation/ablation_summary.json",
           "exists": true,
-          "bytes": 735,
-          "sha256": "2761e1777963473ffa5110ab25e863a6b8ee19ce2004d94ffa938cfa5e0d93fc"
+          "bytes": 755,
+          "sha256": "412b740e36b90f7fbad15c4c4fc46832ab8119d0c84f734b975a5a350836799f"
         }
       },
       "failures": []
@@ -2379,27 +2658,27 @@
       "local": {
         "path": "repo:results/single_episode_diagnostics/timeline_overlay/timeline_overlay.csv",
         "exists": true,
-        "bytes": 293713,
-        "sha256": "eebd53caf853aac5da51adea285e6623dd00860d23d3bc5703169ed6c53d0405"
+        "bytes": 293339,
+        "sha256": "b9fbdb43709e1c4fcea2632c3a8d157297009d82a11c3a3331de1eb92ae17bea"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/single_episode_diagnostics/timeline_overlay/timeline_overlay.csv",
           "exists": true,
-          "bytes": 293713,
-          "sha256": "eebd53caf853aac5da51adea285e6623dd00860d23d3bc5703169ed6c53d0405"
+          "bytes": 293339,
+          "sha256": "b9fbdb43709e1c4fcea2632c3a8d157297009d82a11c3a3331de1eb92ae17bea"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:results/single_episode_diagnostics/timeline_overlay/timeline_overlay.csv",
           "exists": true,
-          "bytes": 293713,
-          "sha256": "eebd53caf853aac5da51adea285e6623dd00860d23d3bc5703169ed6c53d0405"
+          "bytes": 293339,
+          "sha256": "b9fbdb43709e1c4fcea2632c3a8d157297009d82a11c3a3331de1eb92ae17bea"
         },
         "hf_model": {
           "path": "hf_model:results/single_episode_diagnostics/timeline_overlay/timeline_overlay.csv",
           "exists": true,
-          "bytes": 293713,
-          "sha256": "eebd53caf853aac5da51adea285e6623dd00860d23d3bc5703169ed6c53d0405"
+          "bytes": 293339,
+          "sha256": "b9fbdb43709e1c4fcea2632c3a8d157297009d82a11c3a3331de1eb92ae17bea"
         }
       },
       "failures": []
@@ -2410,27 +2689,27 @@
       "local": {
         "path": "repo:results/single_episode_diagnostics/alignment_stress/alignment_shift_metrics.csv",
         "exists": true,
-        "bytes": 8203,
-        "sha256": "8b0026b472a0c0fd8c6eb5c9c307b41ac7ed71fe0ae3d52c8b18dfd5721f0ef1"
+        "bytes": 9654,
+        "sha256": "7bb0b49d7b136af9f34a1430c34f9a2c8e2add793581257f82cc0469cb4acd76"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/single_episode_diagnostics/alignment_stress/alignment_shift_metrics.csv",
           "exists": true,
-          "bytes": 8203,
-          "sha256": "8b0026b472a0c0fd8c6eb5c9c307b41ac7ed71fe0ae3d52c8b18dfd5721f0ef1"
+          "bytes": 9654,
+          "sha256": "7bb0b49d7b136af9f34a1430c34f9a2c8e2add793581257f82cc0469cb4acd76"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:results/single_episode_diagnostics/alignment_stress/alignment_shift_metrics.csv",
           "exists": true,
-          "bytes": 8203,
-          "sha256": "8b0026b472a0c0fd8c6eb5c9c307b41ac7ed71fe0ae3d52c8b18dfd5721f0ef1"
+          "bytes": 9654,
+          "sha256": "7bb0b49d7b136af9f34a1430c34f9a2c8e2add793581257f82cc0469cb4acd76"
         },
         "hf_model": {
           "path": "hf_model:results/single_episode_diagnostics/alignment_stress/alignment_shift_metrics.csv",
           "exists": true,
-          "bytes": 8203,
-          "sha256": "8b0026b472a0c0fd8c6eb5c9c307b41ac7ed71fe0ae3d52c8b18dfd5721f0ef1"
+          "bytes": 9654,
+          "sha256": "7bb0b49d7b136af9f34a1430c34f9a2c8e2add793581257f82cc0469cb4acd76"
         }
       },
       "failures": []
@@ -2442,26 +2721,26 @@
         "path": "repo:results/single_episode_diagnostics/alignment_stress/alignment_stress_summary.json",
         "exists": true,
         "bytes": 332,
-        "sha256": "2a467dae8e2f1606d77eccc8d1cbeb11507d462f95291d52b49d1057b7db9f14"
+        "sha256": "43b03709b7735a20233fa0cd20743b323b2de96e36e50019c4d4aa26e470d2ab"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/single_episode_diagnostics/alignment_stress/alignment_stress_summary.json",
           "exists": true,
           "bytes": 332,
-          "sha256": "2a467dae8e2f1606d77eccc8d1cbeb11507d462f95291d52b49d1057b7db9f14"
+          "sha256": "43b03709b7735a20233fa0cd20743b323b2de96e36e50019c4d4aa26e470d2ab"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:results/single_episode_diagnostics/alignment_stress/alignment_stress_summary.json",
           "exists": true,
           "bytes": 332,
-          "sha256": "2a467dae8e2f1606d77eccc8d1cbeb11507d462f95291d52b49d1057b7db9f14"
+          "sha256": "43b03709b7735a20233fa0cd20743b323b2de96e36e50019c4d4aa26e470d2ab"
         },
         "hf_model": {
           "path": "hf_model:results/single_episode_diagnostics/alignment_stress/alignment_stress_summary.json",
           "exists": true,
           "bytes": 332,
-          "sha256": "2a467dae8e2f1606d77eccc8d1cbeb11507d462f95291d52b49d1057b7db9f14"
+          "sha256": "43b03709b7735a20233fa0cd20743b323b2de96e36e50019c4d4aa26e470d2ab"
         }
       },
       "failures": []
@@ -2472,27 +2751,27 @@
       "local": {
         "path": "repo:QUALITY_GATES.md",
         "exists": true,
-        "bytes": 4919,
-        "sha256": "7138e99b116c44f128cd2f749e9d7427e496cb7596d30d54dd44af79be80df81"
+        "bytes": 4901,
+        "sha256": "7426ef148491d7288161a8d13cf31e81d5351ab1c920614a59b06888889c97d3"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:QUALITY_GATES.md",
           "exists": true,
-          "bytes": 4919,
-          "sha256": "7138e99b116c44f128cd2f749e9d7427e496cb7596d30d54dd44af79be80df81"
+          "bytes": 4901,
+          "sha256": "7426ef148491d7288161a8d13cf31e81d5351ab1c920614a59b06888889c97d3"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:QUALITY_GATES.md",
           "exists": true,
-          "bytes": 4919,
-          "sha256": "7138e99b116c44f128cd2f749e9d7427e496cb7596d30d54dd44af79be80df81"
+          "bytes": 4901,
+          "sha256": "7426ef148491d7288161a8d13cf31e81d5351ab1c920614a59b06888889c97d3"
         },
         "hf_model": {
           "path": "hf_model:QUALITY_GATES.md",
           "exists": true,
-          "bytes": 4919,
-          "sha256": "7138e99b116c44f128cd2f749e9d7427e496cb7596d30d54dd44af79be80df81"
+          "bytes": 4901,
+          "sha256": "7426ef148491d7288161a8d13cf31e81d5351ab1c920614a59b06888889c97d3"
         }
       },
       "failures": []
@@ -2503,27 +2782,27 @@
       "local": {
         "path": "repo:EVALUATION_PROTOCOL.md",
         "exists": true,
-        "bytes": 5938,
-        "sha256": "b170463965704d181db481e0928c5f6fac3a4b626ebc424ec4e977ddead9c0fa"
+        "bytes": 5914,
+        "sha256": "fbb265d5df198c89a096b84d41a877183da03841a1360a6adb7d99fd72f92a29"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:EVALUATION_PROTOCOL.md",
           "exists": true,
-          "bytes": 5938,
-          "sha256": "b170463965704d181db481e0928c5f6fac3a4b626ebc424ec4e977ddead9c0fa"
+          "bytes": 5914,
+          "sha256": "fbb265d5df198c89a096b84d41a877183da03841a1360a6adb7d99fd72f92a29"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:EVALUATION_PROTOCOL.md",
           "exists": true,
-          "bytes": 5938,
-          "sha256": "b170463965704d181db481e0928c5f6fac3a4b626ebc424ec4e977ddead9c0fa"
+          "bytes": 5914,
+          "sha256": "fbb265d5df198c89a096b84d41a877183da03841a1360a6adb7d99fd72f92a29"
         },
         "hf_model": {
           "path": "hf_model:EVALUATION_PROTOCOL.md",
           "exists": true,
-          "bytes": 5938,
-          "sha256": "b170463965704d181db481e0928c5f6fac3a4b626ebc424ec4e977ddead9c0fa"
+          "bytes": 5914,
+          "sha256": "fbb265d5df198c89a096b84d41a877183da03841a1360a6adb7d99fd72f92a29"
         }
       },
       "failures": []
@@ -2535,26 +2814,57 @@
         "path": "repo:FIGURE_INDEX.md",
         "exists": true,
         "bytes": 4712,
-        "sha256": "bf180bd0560c785c85c59f7ad5ad002c1f41ef3b2eda9c604207f39998e4c149"
+        "sha256": "bc1e1b695649270c64f654aa260c9d52cfc372a765ca6c993ef07145e8050f6e"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:FIGURE_INDEX.md",
           "exists": true,
           "bytes": 4712,
-          "sha256": "bf180bd0560c785c85c59f7ad5ad002c1f41ef3b2eda9c604207f39998e4c149"
+          "sha256": "bc1e1b695649270c64f654aa260c9d52cfc372a765ca6c993ef07145e8050f6e"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:FIGURE_INDEX.md",
           "exists": true,
           "bytes": 4712,
-          "sha256": "bf180bd0560c785c85c59f7ad5ad002c1f41ef3b2eda9c604207f39998e4c149"
+          "sha256": "bc1e1b695649270c64f654aa260c9d52cfc372a765ca6c993ef07145e8050f6e"
         },
         "hf_model": {
           "path": "hf_model:FIGURE_INDEX.md",
           "exists": true,
           "bytes": 4712,
-          "sha256": "bf180bd0560c785c85c59f7ad5ad002c1f41ef3b2eda9c604207f39998e4c149"
+          "sha256": "bc1e1b695649270c64f654aa260c9d52cfc372a765ca6c993ef07145e8050f6e"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "docs/FOUNDATION_MODEL_PLAN.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:FOUNDATION_MODEL_PLAN.md",
+        "exists": true,
+        "bytes": 6545,
+        "sha256": "f7d35f38db371d52b5bb92e59b951b93946c425fe12bf283f69f869a45a5c60f"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:FOUNDATION_MODEL_PLAN.md",
+          "exists": true,
+          "bytes": 6545,
+          "sha256": "f7d35f38db371d52b5bb92e59b951b93946c425fe12bf283f69f869a45a5c60f"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:FOUNDATION_MODEL_PLAN.md",
+          "exists": true,
+          "bytes": 6545,
+          "sha256": "f7d35f38db371d52b5bb92e59b951b93946c425fe12bf283f69f869a45a5c60f"
+        },
+        "hf_model": {
+          "path": "hf_model:FOUNDATION_MODEL_PLAN.md",
+          "exists": true,
+          "bytes": 6545,
+          "sha256": "f7d35f38db371d52b5bb92e59b951b93946c425fe12bf283f69f869a45a5c60f"
         }
       },
       "failures": []
@@ -2565,27 +2875,27 @@
       "local": {
         "path": "repo:PROJECT_BRIEF.md",
         "exists": true,
-        "bytes": 2527,
-        "sha256": "0f341ae943873733236b9fefea9a9848c93a64a2e9bb333d1ab697c71f928f15"
+        "bytes": 2516,
+        "sha256": "ace07e60bab3d17d17d2af32b9843582adfbb8d0abe5f7c0489c1e9502eab2f8"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:PROJECT_BRIEF.md",
           "exists": true,
-          "bytes": 2527,
-          "sha256": "0f341ae943873733236b9fefea9a9848c93a64a2e9bb333d1ab697c71f928f15"
+          "bytes": 2516,
+          "sha256": "ace07e60bab3d17d17d2af32b9843582adfbb8d0abe5f7c0489c1e9502eab2f8"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:PROJECT_BRIEF.md",
           "exists": true,
-          "bytes": 2527,
-          "sha256": "0f341ae943873733236b9fefea9a9848c93a64a2e9bb333d1ab697c71f928f15"
+          "bytes": 2516,
+          "sha256": "ace07e60bab3d17d17d2af32b9843582adfbb8d0abe5f7c0489c1e9502eab2f8"
         },
         "hf_model": {
           "path": "hf_model:PROJECT_BRIEF.md",
           "exists": true,
-          "bytes": 2527,
-          "sha256": "0f341ae943873733236b9fefea9a9848c93a64a2e9bb333d1ab697c71f928f15"
+          "bytes": 2516,
+          "sha256": "ace07e60bab3d17d17d2af32b9843582adfbb8d0abe5f7c0489c1e9502eab2f8"
         }
       },
       "failures": []
@@ -2627,27 +2937,27 @@
       "local": {
         "path": "repo:RESEARCH_ROADMAP.md",
         "exists": true,
-        "bytes": 5051,
-        "sha256": "1d640dbc0bcba26d72dbedd4c2dacb04e38bc7c3b11ac79f57b52c2a6e8caec3"
+        "bytes": 6645,
+        "sha256": "b8bebfded058cc2e3f9c918d99600c878f015da08861801fb3ebae73b85fdc57"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 5051,
-          "sha256": "1d640dbc0bcba26d72dbedd4c2dacb04e38bc7c3b11ac79f57b52c2a6e8caec3"
+          "bytes": 6645,
+          "sha256": "b8bebfded058cc2e3f9c918d99600c878f015da08861801fb3ebae73b85fdc57"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 5051,
-          "sha256": "1d640dbc0bcba26d72dbedd4c2dacb04e38bc7c3b11ac79f57b52c2a6e8caec3"
+          "bytes": 6645,
+          "sha256": "b8bebfded058cc2e3f9c918d99600c878f015da08861801fb3ebae73b85fdc57"
         },
         "hf_model": {
           "path": "hf_model:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 5051,
-          "sha256": "1d640dbc0bcba26d72dbedd4c2dacb04e38bc7c3b11ac79f57b52c2a6e8caec3"
+          "bytes": 6645,
+          "sha256": "b8bebfded058cc2e3f9c918d99600c878f015da08861801fb3ebae73b85fdc57"
         }
       },
       "failures": []
@@ -2658,27 +2968,27 @@
       "local": {
         "path": "repo:PROJECT_STATUS.md",
         "exists": true,
-        "bytes": 5385,
-        "sha256": "353e177fac7a2009c475b8ae833117ae4cf68ca69307d9e1c292b6e13a18be61"
+        "bytes": 6710,
+        "sha256": "22c3ba04620eb003b6ba206f55cfe02c723c41b0c972dcd5a03f0aade1e43fda"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 5385,
-          "sha256": "353e177fac7a2009c475b8ae833117ae4cf68ca69307d9e1c292b6e13a18be61"
+          "bytes": 6710,
+          "sha256": "22c3ba04620eb003b6ba206f55cfe02c723c41b0c972dcd5a03f0aade1e43fda"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 5385,
-          "sha256": "353e177fac7a2009c475b8ae833117ae4cf68ca69307d9e1c292b6e13a18be61"
+          "bytes": 6710,
+          "sha256": "22c3ba04620eb003b6ba206f55cfe02c723c41b0c972dcd5a03f0aade1e43fda"
         },
         "hf_model": {
           "path": "hf_model:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 5385,
-          "sha256": "353e177fac7a2009c475b8ae833117ae4cf68ca69307d9e1c292b6e13a18be61"
+          "bytes": 6710,
+          "sha256": "22c3ba04620eb003b6ba206f55cfe02c723c41b0c972dcd5a03f0aade1e43fda"
         }
       },
       "failures": []
@@ -2720,27 +3030,27 @@
       "local": {
         "path": "repo:RESEARCH_TAKEAWAYS.md",
         "exists": true,
-        "bytes": 3833,
-        "sha256": "86e2cbbb820a0fd017c709759b952336364316ff166985e5de5f0c4666b2c55c"
+        "bytes": 4889,
+        "sha256": "1f688343561501898fdc615f8ac67ac1ae64fc2acc4290b3a17a5b4f0c732b99"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:RESEARCH_TAKEAWAYS.md",
           "exists": true,
-          "bytes": 3833,
-          "sha256": "86e2cbbb820a0fd017c709759b952336364316ff166985e5de5f0c4666b2c55c"
+          "bytes": 4889,
+          "sha256": "1f688343561501898fdc615f8ac67ac1ae64fc2acc4290b3a17a5b4f0c732b99"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:RESEARCH_TAKEAWAYS.md",
           "exists": true,
-          "bytes": 3833,
-          "sha256": "86e2cbbb820a0fd017c709759b952336364316ff166985e5de5f0c4666b2c55c"
+          "bytes": 4889,
+          "sha256": "1f688343561501898fdc615f8ac67ac1ae64fc2acc4290b3a17a5b4f0c732b99"
         },
         "hf_model": {
           "path": "hf_model:RESEARCH_TAKEAWAYS.md",
           "exists": true,
-          "bytes": 3833,
-          "sha256": "86e2cbbb820a0fd017c709759b952336364316ff166985e5de5f0c4666b2c55c"
+          "bytes": 4889,
+          "sha256": "1f688343561501898fdc615f8ac67ac1ae64fc2acc4290b3a17a5b4f0c732b99"
         }
       },
       "failures": []
@@ -2752,26 +3062,26 @@
         "path": "repo:SOURCE_ALIGNMENT_AUDIT.md",
         "exists": true,
         "bytes": 2309,
-        "sha256": "26f8234b6a537ee5961a140e38ae78d9fedda4158483703d7853b6d7791cd9cb"
+        "sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:SOURCE_ALIGNMENT_AUDIT.md",
           "exists": true,
           "bytes": 2309,
-          "sha256": "26f8234b6a537ee5961a140e38ae78d9fedda4158483703d7853b6d7791cd9cb"
+          "sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:SOURCE_ALIGNMENT_AUDIT.md",
           "exists": true,
           "bytes": 2309,
-          "sha256": "26f8234b6a537ee5961a140e38ae78d9fedda4158483703d7853b6d7791cd9cb"
+          "sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
         },
         "hf_model": {
           "path": "hf_model:SOURCE_ALIGNMENT_AUDIT.md",
           "exists": true,
           "bytes": 2309,
-          "sha256": "26f8234b6a537ee5961a140e38ae78d9fedda4158483703d7853b6d7791cd9cb"
+          "sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
         }
       },
       "failures": []
@@ -2782,27 +3092,27 @@
       "local": {
         "path": "repo:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
         "exists": true,
-        "bytes": 10597,
-        "sha256": "3103ed45295e75cca79df6b92709a286989355772fde0f2e573ba1a2bd12d499"
+        "bytes": 10625,
+        "sha256": "9e51442cb13cb4074d73fb212b49c56e6e4ae6dea10fd135a4020f7488e49be0"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
           "exists": true,
-          "bytes": 10597,
-          "sha256": "3103ed45295e75cca79df6b92709a286989355772fde0f2e573ba1a2bd12d499"
+          "bytes": 10625,
+          "sha256": "9e51442cb13cb4074d73fb212b49c56e6e4ae6dea10fd135a4020f7488e49be0"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
           "exists": true,
-          "bytes": 10597,
-          "sha256": "3103ed45295e75cca79df6b92709a286989355772fde0f2e573ba1a2bd12d499"
+          "bytes": 10625,
+          "sha256": "9e51442cb13cb4074d73fb212b49c56e6e4ae6dea10fd135a4020f7488e49be0"
         },
         "hf_model": {
           "path": "hf_model:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
           "exists": true,
-          "bytes": 10597,
-          "sha256": "3103ed45295e75cca79df6b92709a286989355772fde0f2e573ba1a2bd12d499"
+          "bytes": 10625,
+          "sha256": "9e51442cb13cb4074d73fb212b49c56e6e4ae6dea10fd135a4020f7488e49be0"
         }
       },
       "failures": []

data/publication_audit.json

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-03T07:49:54+00:00",
+  "generated_at_utc": "2026-06-03T19:29:55+00:00",
   "checks": [
     {
       "name": "required_publication_assets_present",
@@ -182,44 +182,44 @@
     "github_repo": {
       "root": "repo",
       "exists": true,
-      "file_count": 355,
-      "text_file_count": 291,
+      "file_count": 380,
+      "text_file_count": 315,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
-        "bytes": 52601010
+        "bytes": 55702978
       },
       "violations": []
     },
     "hf_space_bundle": {
       "root": "hf_publish/space",
       "exists": true,
-      "file_count": 137,
-      "text_file_count": 110,
+      "file_count": 312,
+      "text_file_count": 247,
       "largest_file": {
-        "path": "data/single_episode_explorer.json",
-        "bytes": 4101241
+        "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
+        "bytes": 55702978
       },
       "violations": []
     },
     "hf_artifact_bundle": {
       "root": "hf_publish/artifacts",
       "exists": true,
-      "file_count": 384,
-      "text_file_count": 300,
+      "file_count": 410,
+      "text_file_count": 325,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
-        "bytes": 52601010
+        "bytes": 55702978
       },
       "violations": []
     },
     "hf_model_bundle": {
       "root": "hf_publish/model",
       "exists": true,
-      "file_count": 569,
-      "text_file_count": 453,
+      "file_count": 635,
+      "text_file_count": 513,
       "largest_file": {
-        "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
-        "bytes": 52601010
+        "path": "artifacts/episode_task_suite/modality_reconstruction/predictions.npz",
+        "bytes": 55702978
       },
       "violations": []
     }

docs/data/artifact_index.json

@@ -1,6 +1,6 @@
 {
   "title": "Ropedia Xperience-10M Task Suite Artifact Index",
-  "generated_at_utc": "2026-06-03T16:58:00+00:00",
+  "generated_at_utc": "2026-06-03T18:11:35+00:00",
   "status": "pass",
   "artifact_count": 72,
   "missing": [],
@@ -40,8 +40,8 @@
       "surface": "repo_hf",
       "shows": "Gives first-pass readers a concise project shape before the detailed artifact trail.",
       "exists": true,
-      "bytes": 2494,
-      "sha256": "54c6906103f6d7f45398cd638730c3eb9813d6ca4a807920d6edba4b12f446e6"
+      "bytes": 2516,
+      "sha256": "ace07e60bab3d17d17d2af32b9843582adfbb8d0abe5f7c0489c1e9502eab2f8"
     },
     {
       "id": "project_brief_json",
@@ -51,8 +51,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable first-reader project brief for the website and Hugging Face mirrors.",
       "exists": true,
-      "bytes": 2563,
-      "sha256": "c8df06022de6a0d51be993175ac63764a607f48c944f1d9451c5bbeebf6cef54"
+      "bytes": 2585,
+      "sha256": "f0e2e7853b86cdcd2f06269d75f19a96432a782efdf99c08aca7d689fd2cd0a1"
     },
     {
       "id": "project_status",
@@ -62,8 +62,8 @@
       "surface": "repo_hf",
       "shows": "Gives a compact current-state table for first-pass readers.",
       "exists": true,
-      "bytes": 6721,
-      "sha256": "166795e9f5986015f7a9e74934fbd878f5dcc7cbac79d49483bd514feacbfde8"
+      "bytes": 6648,
+      "sha256": "c873c0361ee1b83a835f1112529a557e2033959d70a9bfe726a66cdd120a9047"
     },
     {
       "id": "project_status_json",
@@ -73,8 +73,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable copy of the current project status for website and HF mirrors.",
       "exists": true,
-      "bytes": 9006,
-      "sha256": "38d289ce14e7b4f0998ce156718662f2298ee3aa4e77782a40316ad33aefaefb"
+      "bytes": 8868,
+      "sha256": "ea2deebfe7dd2ee7cb141623f4c6c1d4d287bb94a1c642609b854e934d5d701a"
     },
     {
       "id": "research_roadmap",
@@ -84,8 +84,8 @@
       "surface": "repo_hf",
       "shows": "Defines the staged path from public-sample task development to multi-episode held-out evaluation and larger omni-model extensions.",
       "exists": true,
-      "bytes": 6677,
-      "sha256": "590702b8000bd143fdbcf65104b0cad835e6e4c79754c28c31727f1bdf43ae79"
+      "bytes": 6645,
+      "sha256": "b8bebfded058cc2e3f9c918d99600c878f015da08861801fb3ebae73b85fdc57"
     },
     {
       "id": "research_roadmap_json",
@@ -95,8 +95,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable staged roadmap for the website and Hugging Face mirrors.",
       "exists": true,
-      "bytes": 5752,
-      "sha256": "0510e919ad9ffa3aebe052273e90133a57572c3b21b991db1ff28de83061f6ca"
+      "bytes": 5726,
+      "sha256": "903ec0d6104c092fa0356c7741134fee4a1c78f3b0517001c49ce11f8077d6e2"
     },
     {
       "id": "foundation_model_plan",
@@ -106,8 +106,8 @@
       "surface": "repo_hf",
       "shows": "Defines the post-data-gate backbone choices: Qwen3-Omni first, Cosmos 3 for world modeling, and VLA/policy models after action-target conversion.",
       "exists": true,
-      "bytes": 6538,
-      "sha256": "15431b3ed368d660e777655d2e4a97f2870997f2ad38b451a73d631b5e431bcf"
+      "bytes": 6545,
+      "sha256": "f7d35f38db371d52b5bb92e59b951b93946c425fe12bf283f69f869a45a5c60f"
     },
     {
       "id": "foundation_model_plan_json",
@@ -117,8 +117,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable foundation-model selection matrix with source links, entry conditions, and evaluation additions.",
       "exists": true,
-      "bytes": 8883,
-      "sha256": "0465df50071ab4c8e46acafdd2fa1653f23dacd33a1e1e1d65a571c82348ddbc"
+      "bytes": 8880,
+      "sha256": "a11476cc697e7bc2c29b3db569d059c8f536025fdd0b47e49f060a88664fd450"
     },
     {
       "id": "evidence_contract",
@@ -128,8 +128,8 @@
       "surface": "repo",
       "shows": "Defines the implemented scope, setup-stage items, and multi-episode prerequisites.",
       "exists": true,
-      "bytes": 11172,
-      "sha256": "7197fe7e42ea4cbe200b203871d0db745c201c37eef088243aa21ac4a72fe782"
+      "bytes": 11287,
+      "sha256": "46b03a5c31ec31476c0476b11bc72375d786a5e8ea11fcba81e1500a0e50b9b8"
     },
     {
       "id": "project_packet",
@@ -139,8 +139,8 @@
       "surface": "website_hf",
       "shows": "Gives a short project path with scope status and public surfaces.",
       "exists": true,
-      "bytes": 7659,
-      "sha256": "7d9f5ecc8adf9b3913e99a8aec28372529e63c105739fa53b3f8a7a9470470c2"
+      "bytes": 7553,
+      "sha256": "2f2d227bf3d95eeba7e4f5133d37fbe62910cd373aa3431d9b960914a590e149"
     },
     {
       "id": "artifact_guide",
@@ -150,8 +150,8 @@
       "surface": "repo_hf",
       "shows": "Gives the human-readable map from project scope to data, tasks, platform mirrors, and scale-up status.",
       "exists": true,
-      "bytes": 16974,
-      "sha256": "e456d291614db14a2cb6a91cabf1b044f47c518d39d7d701eb191f9434f2c157"
+      "bytes": 16876,
+      "sha256": "6b392ff36290f8eefab393218c17aff4d90ba28457324089f78e7a4882b17dbc"
     },
     {
       "id": "official_dataset_card_alignment",
@@ -161,8 +161,8 @@
       "surface": "repo_hf",
       "shows": "Aligns public dataset wording with the official gated Xperience-10M card, public sample card, HF API metadata, and current project coverage.",
       "exists": true,
-      "bytes": 10719,
-      "sha256": "a0cc66e3a52d253dad4b8b19396cd3a8db14bf1835034c6c0aea6c5248fe4037"
+      "bytes": 10625,
+      "sha256": "9e51442cb13cb4074d73fb212b49c56e6e4ae6dea10fd135a4020f7488e49be0"
     },
     {
       "id": "official_dataset_card_alignment_json",
@@ -172,8 +172,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable upstream dataset-card, sample-card, and HF API alignment facts for website and HF mirrors.",
       "exists": true,
-      "bytes": 7573,
-      "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7"
+      "bytes": 7559,
+      "sha256": "24ea3fc9e1ae8f698ca192ae1b8511dfe1129b807af237f41382e0644e5b58af"
     },
     {
       "id": "source_alignment",
@@ -195,7 +195,7 @@
       "shows": "Machine-readable source-alignment pass/fail check for repo, website, and HF surfaces.",
       "exists": true,
       "bytes": 4432,
-      "sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
+      "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
     },
     {
       "id": "source_alignment_validator",
@@ -205,8 +205,8 @@
       "surface": "repo_hf",
       "shows": "Regenerates the source-alignment report from committed facts and public card text.",
       "exists": true,
-      "bytes": 15807,
-      "sha256": "66a9d0eebb4800be5d65f982ef13502b37522cd2a9eae0686c63efc121a52b0a"
+      "bytes": 15831,
+      "sha256": "6f48190ca77dd6ba9dd2f821d5ace7e33bd9b4d9d76adb7237cc51ddb88de239"
     },
     {
       "id": "hf_publisher",
@@ -238,8 +238,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable protocol generated from committed task metrics for website and HF mirrors.",
       "exists": true,
-      "bytes": 13644,
-      "sha256": "69281b1727d3de7a734fe89846e9881b02fa1df4108f4ae39946acbcdf6ed8bf"
+      "bytes": 13625,
+      "sha256": "33a9589ad613bf524c5db932660b810cb724c8765d8bfa65a9996b01286e39fa"
     },
     {
       "id": "evaluation_protocol_builder",
@@ -249,8 +249,8 @@
       "surface": "repo_hf",
       "shows": "Regenerates the protocol from committed summary metrics and task artifacts.",
       "exists": true,
-      "bytes": 16163,
-      "sha256": "b352798b50212dac792855313bdc0ad1e5e2ef835de1ff25b03d52afd1d33f03"
+      "bytes": 16139,
+      "sha256": "e2aaac38f451402971e03c004d32d5c5f8a05d67dc1820d6bf9b12d1da1a292f"
     },
     {
       "id": "research_takeaways",
@@ -260,8 +260,8 @@
       "surface": "repo_hf",
       "shows": "Summarizes the main research lessons from committed metrics and identifies which experiments need held-out episodes.",
       "exists": true,
-      "bytes": 4893,
-      "sha256": "16a002167d0702ea7e85ad1a494800daad4ff080c3751e078275f7c196721046"
+      "bytes": 4889,
+      "sha256": "1f688343561501898fdc615f8ac67ac1ae64fc2acc4290b3a17a5b4f0c732b99"
     },
     {
       "id": "research_takeaways_json",
@@ -271,8 +271,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable result interpretation for the website, HF cards, and mirror checks.",
       "exists": true,
-      "bytes": 6814,
-      "sha256": "538ec80950dbf5d77cc6ab2bd5a96a3af191b56a257b895e48b6cf28deb0d044"
+      "bytes": 6821,
+      "sha256": "509d8690a4854097ea45af635dafa0231efa11838f1bb1f9cb338e99009a0206"
     },
     {
       "id": "research_takeaways_builder",
@@ -282,19 +282,19 @@
       "surface": "repo_hf",
       "shows": "Regenerates the research takeaways from committed summary metrics and task result artifacts.",
       "exists": true,
-      "bytes": 13304,
-      "sha256": "135f555b17aae5dc5965ec88c3db6e49622a480467909ffa10d53278d4ea054b"
+      "bytes": 13285,
+      "sha256": "19baa1dbe3e4454e040bb1b2fc711e1dd5ce5280ee19bfc01b01cca854006d38"
     },
     {
       "id": "audio_ablation_script",
-      "title": "Audio ablation and raw-audio upgrade script",
+      "title": "Audio contribution script",
       "path": "scripts/audio_ablation_and_raw_upgrade.py",
       "kind": "result_interpretation",
       "surface": "repo_hf",
-      "shows": "Measures current AAC audio contribution and a raw log-mel audio feature replacement across all 12 task contracts.",
+      "shows": "Measures audio contribution variants across all 12 task contracts.",
       "exists": true,
-      "bytes": 43192,
-      "sha256": "73db45a3ddb50b60f61ca5e55bdf2008e48c268681bfcdc95d35bb56e48f89e3"
+      "bytes": 43144,
+      "sha256": "f7e3a38ec906dac7ca55b13c49720bd41ed89a1fd994c7d54730a4de5dfd1b59"
     },
     {
       "id": "audio_ablation_summary",
@@ -302,7 +302,7 @@
       "path": "results/audio_ablation/audio_ablation_summary.json",
       "kind": "metrics_source",
       "surface": "repo_hf",
-      "shows": "Stores per-task audio deltas for all current features, no-audio, handcrafted-audio-only, raw-audio-only, raw replacement, and all-plus-raw variants.",
+      "shows": "Stores per-task audio deltas for all current features, no-audio, audio-only, alternate-audio-only, replacement, and all-plus-alternate variants.",
       "exists": true,
       "bytes": 9735,
       "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
@@ -313,7 +313,7 @@
       "path": "results/audio_ablation/AUDIO_ABLATION_SUMMARY.md",
       "kind": "result_interpretation",
       "surface": "repo_hf",
-      "shows": "Human-readable table showing the measured audio contribution and raw-audio replacement delta for every task.",
+      "shows": "Human-readable table showing the measured audio contribution and alternate-representation delta for every task.",
       "exists": true,
       "bytes": 2166,
       "sha256": "9318bc55b1b051c384ce2fd6e7cd0a56becfa00185a0b597fdc474a40b1f17f9"
@@ -326,8 +326,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable audio ablation summary mirrored into the static website and Hugging Face bundles.",
       "exists": true,
-      "bytes": 9735,
-      "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
+      "bytes": 9701,
+      "sha256": "7debd9e67b7df09322d743193bf8b785277a992c9e8c82c08c7a36c4e066e6de"
     },
     {
       "id": "audio_ablation_delta_chart",
@@ -414,8 +414,8 @@
       "surface": "repo_hf",
       "shows": "Lists the automated and post-publish checks used to keep the release current.",
       "exists": true,
-      "bytes": 4919,
-      "sha256": "7138e99b116c44f128cd2f749e9d7427e496cb7596d30d54dd44af79be80df81"
+      "bytes": 4901,
+      "sha256": "7426ef148491d7288161a8d13cf31e81d5351ab1c920614a59b06888889c97d3"
     },
     {
       "id": "quality_gate_manifest",
@@ -425,8 +425,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable release-check summary for validators, mirrors, and public project surfaces.",
       "exists": true,
-      "bytes": 8147,
-      "sha256": "1d4d45937e7d90ee5d54125e4d5d7b5d5eee4878251b52152e59a34efc557c04"
+      "bytes": 8118,
+      "sha256": "9c40b314c53d8406e52288656e699ad707a7cc2aa5161bdb5117f1a5b84e4ea2"
     },
     {
       "id": "public_surface_qa",
@@ -529,7 +529,7 @@
       "volatile": true,
       "shows": "Records the last live GitHub/HF URL verification after upload.",
       "exists": true,
-      "bytes": 68744,
+      "bytes": 68749,
       "hash_policy": "existence_and_size_only"
     },
     {
@@ -540,8 +540,8 @@
       "surface": "repo",
       "shows": "Fetches the published GitHub/HF URLs and compares live hashes and public-card markers against the release assets.",
       "exists": true,
-      "bytes": 34767,
-      "sha256": "6a1cb0531719888de6bbeb7a2628058f6e67c92df6314c4ac3952af2c17f126b"
+      "bytes": 34779,
+      "sha256": "73c45e1ae1b7509b9f2c54cf42424c9ac0aab8a9e54be58fc488eaa2a696acc7"
     },
     {
       "id": "reproducibility_contract",
@@ -551,8 +551,8 @@
       "surface": "repo_hf",
       "shows": "Defines public reproduction commands, expected outputs, and non-reproducible scale-up boundaries.",
       "exists": true,
-      "bytes": 6312,
-      "sha256": "99a2df715ce481f49991de620f20e5093f37767fa36122c1505c74ee6ea92e3c"
+      "bytes": 6346,
+      "sha256": "48d82bbdfeba88a63100f55e19ef7e1392c0e633b5f3f0c648444f9ed1c96e02"
     },
     {
       "id": "reproducibility_matrix",
@@ -562,8 +562,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable reproduction steps with expected artifacts and public boundaries.",
       "exists": true,
-      "bytes": 5197,
-      "sha256": "c3d353084ee641ff3d1369dd911577ff7a1b5d7b2faf841faa12caa69dfa5ed0"
+      "bytes": 5229,
+      "sha256": "928eb95b95d4028a7f21135584c37ff141833e26178cc918a874a6448f8c5ded"
     },
     {
       "id": "artifact_index_builder",
@@ -573,8 +573,8 @@
       "surface": "repo_hf",
       "shows": "Generates the selective artifact catalog from local files.",
       "exists": true,
-      "bytes": 26652,
-      "sha256": "faf1e0b56eb9ba4bc94e04bb1aeb9a392fb4410e8053564dd43bce3d8aacfa4d"
+      "bytes": 26566,
+      "sha256": "982fcc36e6dfda3382662d1f1e1f539f27f45aaddd6503aad77a14145bcc1bf6"
     },
     {
       "id": "publication_audit",
@@ -595,7 +595,7 @@
       "kind": "scale_up_status",
       "surface": "website_hf",
       "volatile": true,
-      "shows": "Records historical 32ep setup paths separately from completed held-out-episode results.",
+      "shows": "Separates setup paths from completed held-out-episode results.",
       "exists": true,
       "bytes": 20066,
       "hash_policy": "existence_and_size_only"
@@ -654,8 +654,8 @@
       "surface": "website_hf",
       "shows": "Mirrors task metrics for the static dashboard.",
       "exists": true,
-      "bytes": 25210,
-      "sha256": "a88302490b9099e7d148a8a6dd0ba6393d7ef26a0ba2c6a0edf552da94b948a8"
+      "bytes": 25198,
+      "sha256": "a5c337bc6be321a9999dd69f0920756ad722d505697827425a64005f05a74699"
     },
     {
       "id": "feature_manifest",
@@ -709,8 +709,8 @@
       "surface": "repo_hf",
       "shows": "Maps the 12 tasks to the four Ropedia research directions as direct/proxy/diagnostic.",
       "exists": true,
-      "bytes": 14429,
-      "sha256": "292ddd47b5f50cc5a6a0d4099466b134774a1ca7610afe50a2a5781c6c181bf9"
+      "bytes": 14414,
+      "sha256": "7215681c55a6739da7f16d833f62fcb8f6d58069840963182e98427eaf0cf654"
     },
     {
       "id": "research_direction_extensions",
@@ -720,8 +720,8 @@
       "surface": "repo_hf",
       "shows": "Stores one coded extension probe per research direction with minimal and neural metrics.",
       "exists": true,
-      "bytes": 11907,
-      "sha256": "aa65022a8a41af4f0bca5197add932c4f3f51b13b8324323280f64457473ff03"
+      "bytes": 11904,
+      "sha256": "b72bb62f95d27de296ae526568c55bca3aeb9afe687e57bbe652ca74a32c93d7"
     },
     {
       "id": "task_walkthroughs",
@@ -731,8 +731,8 @@
       "surface": "repo_hf",
       "shows": "Explains every task with case study, input, process modules, output, and limitation.",
       "exists": true,
-      "bytes": 15449,
-      "sha256": "9309a4f6da85c43c9d87be6ed0458c3cbc9c01b664da6610cbbb2392d282302e"
+      "bytes": 15436,
+      "sha256": "de8607242ca8c5c1c632f7175eeffd0c009536eb050614bfae2ff561e14cd92d"
     },
     {
       "id": "task_suite_infographic",
@@ -753,8 +753,8 @@
       "surface": "website_hf",
       "shows": "Documents the seven public-sample modality cards and their derived thumbnail assets.",
       "exists": true,
-      "bytes": 3819,
-      "sha256": "b1285148588ef402b70c830b62d66fbbcfa86bc5ffc2ad99e165a1429bb49084"
+      "bytes": 3797,
+      "sha256": "4f12a5471d4fe53b1f5c1ef6f9535cc14974690fa6ffbfbeddc8e1b90f86b281"
     },
     {
       "id": "modality_thumbnails",
@@ -795,10 +795,10 @@
       "path": "results/omni_finetune/DATA_ACCESS_STATUS.md",
       "kind": "scaleup_status",
       "surface": "repo_hf",
-      "shows": "Summarizes the data-access requirement before the 32-episode Qwen3-Omni pilot can run.",
+      "shows": "Summarizes the staging requirement before the held-out Qwen3-Omni pilot can report metrics.",
       "exists": true,
-      "bytes": 2954,
-      "sha256": "6ac14781581f06004bdc2fb0e2425c616419b738e6b858f8611bd19fab80741c"
+      "bytes": 2940,
+      "sha256": "c33c40b2bdda0a057cbfef28e1fc75695e4e7273703fc261137f796fded27d47"
     },
     {
       "id": "multi_episode_access_status",
@@ -808,8 +808,8 @@
       "surface": "repo_hf",
       "shows": "Documents the public multi-episode access status and 32-episode pilot selection.",
       "exists": true,
-      "bytes": 2282,
-      "sha256": "e270cdfa4485458572a40f5c3c4bd598fe53e2b8de60af3ff1364398d96862e2"
+      "bytes": 2285,
+      "sha256": "e05b3396bb7c1254ab516d154f7deca293be188d3a48624862822b2b5e26dc91"
     },
     {
       "id": "citation",

docs/data/audio_ablation_summary.json

@@ -1,5 +1,5 @@
 {
-  "description": "Measured audio ablation and raw log-mel audio upgrade over the single public Xperience-10M sample episode.",
+  "description": "Measured audio contribution variants over the single public Xperience-10M sample episode.",
   "scope": "single public sample episode; chronological split; ridge heads over fixed feature contracts",
   "raw_audio_metadata": {
     "source": "local_public_sample/fisheye_cam0.mp4",
@@ -19,9 +19,9 @@
   "variants": {
     "all_handcrafted_audio": "All Current Features",
     "all_except_audio": "All Except Audio",
-    "handcrafted_audio_only": "Handcrafted AAC Audio Only",
+    "handcrafted_audio_only": "Audio Only",
     "raw_logmel_audio_only": "Raw Log-Mel Audio Only",
-    "replace_handcrafted_with_raw": "Replace AAC Block With Raw Log-Mel",
+    "replace_handcrafted_with_raw": "Audio Representation Replacement",
     "all_plus_raw_logmel": "All Current Features + Raw Log-Mel"
   },
   "task_summaries": [
@@ -220,4 +220,4 @@
     "annotation_source": "local_public_sample/annotation.hdf5",
     "homie_toolkit_available": true
   }
-}
+}

docs/data/evaluation_protocol.json

@@ -28,10 +28,10 @@
     "limitation": "It is still one episode; cross-episode generalization is evaluated in the multi-episode stage."
   },
   "feature_policy": {
-    "input_contract": "8,546-dimensional current feature vector",
+    "input_contract": "8,546-dimensional aligned multimodal window representation",
     "source_manifest": "results/episode_task_suite/feature_manifest.json",
     "normalization": "Scalers are fit on train windows only for the baseline heads.",
-    "audio_status": "AAC audio is extracted from the sample MP4 stream and included in the current feature vector."
+    "audio_status": "Audio is one of the synchronized source modalities in the current task representation."
   },
   "baselines": [
     {
@@ -166,7 +166,7 @@
       "task": "contact_prediction",
       "family": "binary classification",
       "unit": "single window",
-      "input": "non-contact and non-caption feature blocks",
+      "input": "non-contact and non-caption signals",
       "target": "any body contact",
       "primary_metric": "macro_f1",
       "higher_is_better": true,
@@ -185,7 +185,7 @@
       "task": "object_relevance",
       "family": "multi-label classification",
       "unit": "single window",
-      "input": "non-caption feature blocks",
+      "input": "non-caption signals",
       "target": "current relevant object set",
       "primary_metric": "micro_f1",
       "higher_is_better": true,
@@ -227,7 +227,7 @@
       "target": "matching depth/video window",
       "primary_metric": "top5_accuracy",
       "higher_is_better": true,
-      "leakage_rule": "Query-side and candidate-side feature blocks are split before projection/ranking.",
+      "leakage_rule": "Query-side and candidate-side signals are split before projection/ranking.",
       "counts": {
         "num_queries": 348,
         "num_train_windows": 813,
@@ -246,7 +246,7 @@
       "target": "depth/video feature vector",
       "primary_metric": "r2",
       "higher_is_better": true,
-      "leakage_rule": "Target feature blocks are excluded from the input side.",
+      "leakage_rule": "Target-side signals are excluded from the input side.",
       "counts": {
         "num_train_windows": 813,
         "num_test_windows": 348
@@ -299,23 +299,23 @@
     "Use chronological train/test splits instead of random window shuffling.",
     "Fit scalers and learned projections on train windows only.",
     "Keep future labels, future mocap, contact labels, object labels, and caption labels on the target side unless a task explicitly treats language as the query.",
-    "For cross-modal tasks, split query-side and candidate-side feature blocks before training and ranking.",
+    "For cross-modal tasks, split query-side and candidate-side signals before training and ranking.",
     "Report unseen test classes when the chronological split exposes labels absent from the train segment."
   ],
   "current_limitations": [
     "Cross-episode generalization is evaluated in the later multi-episode stage.",
     "Feature-vector reconstruction is separate from pixel depth, mesh, NeRF, or Gaussian reconstruction.",
-    "Qwen3-Omni setup artifacts are preparation artifacts until the 32-episode held-out pilot runs.",
-    "Full audio-visual representation learning still needs multi-episode training, but the current baseline vector now includes an extracted AAC audio feature block."
+    "Qwen3-Omni setup artifacts are preparation artifacts until the selected held-out pilot runs.",
+    "Full audio-visual representation learning still needs multi-episode training; the current report includes single-episode audio/no-audio ablations."
   ],
   "scale_up_gate": {
     "required_before_full_omni_pilot": [
-      "at least 32 valid Xperience-10M episodes",
+      "selected staged Xperience-10M episodes",
       "held-out episode split with no train/test episode leakage",
       "manifest, training metadata, progress logs, metrics, predictions, and run report",
       "held-out evaluation on test episodes rather than train windows"
     ],
-    "current_status": "prepared but data-gated",
+    "current_status": "prepared; selected data relay in progress",
     "evidence": [
       "results/omni_finetune/DATA_ACCESS_STATUS.md",
       "results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md"

docs/data/evidence_contract.json

@@ -63,7 +63,7 @@
                 "results/episode_task_suite/feature_manifest.json",
                 "results/episode_task_suite/available_modalities.json"
             ],
-            "boundary": "8,546-d feature vector, including audio_fisheye_cam0_aac decoded from the sample MP4"
+            "boundary": "8,546-dimensional aligned multimodal window representation"
         },
         {
             "id": "evaluation_protocol",
@@ -171,28 +171,28 @@
         "results/omni_finetune/dataset_manifest.json",
         "results/omni_finetune/metrics_eval.json"
       ],
-      "boundary": "one episode and 128 train windows; full metrics require the 32-episode pilot"
+      "boundary": "one episode and 128 train windows; full metrics require completed multi-episode staging and held-out evaluation"
     },
     {
-      "id": "thirty_two_episode_gate",
-      "claim": "The 32-episode LoRA pilot is waiting on gated data access.",
-      "status": "pending_data_access",
+      "id": "multi_episode_staging",
+      "claim": "The Qwen3-Omni LoRA pilot is in multi-episode staging.",
+      "status": "data_staging",
       "evidence": [
         "results/omni_finetune/DATA_ACCESS_STATUS.md",
         "results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md",
         "results/omni_finetune/source_discovery.json"
       ],
-      "boundary": "held-out metrics come after the data gate, manifest construction, training, and test evaluation"
+      "boundary": "full-dataset access is granted; held-out metrics come after selected relay, manifest construction, training, and test evaluation"
     },
     {
       "id": "scale_up_status_check",
-      "claim": "Historical 32ep path strings are tracked as setup-file provenance.",
+      "claim": "Older pilot path strings are tracked as setup-file provenance.",
       "status": "verified",
       "evidence": [
         "scripts/validate_scope_claims.py",
         "docs/data/scope_claims_audit.json"
       ],
-      "boundary": "old run/path identifiers stay separate from completed 32-episode results"
+      "boundary": "run/path identifiers stay separate from completed held-out-episode results"
     },
     {
       "id": "mirror_parity",
@@ -301,7 +301,7 @@
         "docs/data/reproducibility_matrix.json",
         "notes/reproducibility_audit.md"
       ],
-      "boundary": "publicly reproduces the single-episode pipeline, not the gated 32-episode Qwen3-Omni pilot"
+      "boundary": "publicly reproduces the single-episode pipeline; multi-episode Qwen3-Omni metrics are added only after staging and held-out evaluation"
     }
   ]
 }

docs/data/foundation_model_plan.json

@@ -18,14 +18,14 @@
       "family": "Qwen3-Omni",
       "category": "omni_instruction_model",
       "openness": "open_weights_available_from_official_hf_repo",
-      "best_role": "First 32-episode multimodal LoRA pilot and structured task predictor.",
+      "best_role": "First selected-episode multimodal LoRA pilot and structured task predictor.",
       "xperience10m_fit": [
         "RGB/fisheye video, embedded audio, and language prompts can enter directly.",
         "Depth, pose/SLAM, mocap, contacts, and IMU enter through the existing sensor bridge.",
         "Matches current task outputs: labels, structured JSON, captions, and short decisions."
       ],
       "current_decision": "keep_as_first_pilot",
-      "entry_condition": "At least 32 valid episodes staged with held-out episode split.",
+      "entry_condition": "Selected episodes staged with held-out episode split.",
       "public_source": "https://huggingface.co/Qwen/Qwen3-Omni-30B-A3B-Instruct"
     },
     {

docs/data/live_publication_status.json

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Live Publication Status",
   "status": "pass",
-  "checked_at_utc": "2026-06-03T13:53:58+00:00",
+  "checked_at_utc": "2026-06-03T17:18:16+00:00",
   "scope": "Live GitHub Pages, GitHub raw, Hugging Face Space, artifact dataset, and model card mirrors.",
   "hash_groups": [
     {
@@ -58,14 +58,14 @@
         "path": "docs/data/quality_gates.json",
         "exists": true,
         "bytes": 8147,
-        "sha256": "40c9cdc9d5afd15f8c69bda48a0eb36d81df14adff5555de54a481df1baed5ca"
+        "sha256": "1a6b09fdf86ad0ad6aef167de8325670d6643c5d68c35e9846ff85270dcfed7a"
       },
       "mirrors": {
         "github_pages": {
           "ok": true,
           "status_code": 200,
           "bytes": 8147,
-          "sha256": "40c9cdc9d5afd15f8c69bda48a0eb36d81df14adff5555de54a481df1baed5ca",
+          "sha256": "1a6b09fdf86ad0ad6aef167de8325670d6643c5d68c35e9846ff85270dcfed7a",
           "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/quality_gates.json",
           "url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/quality_gates.json"
         },
@@ -73,7 +73,7 @@
           "ok": true,
           "status_code": 200,
           "bytes": 8147,
-          "sha256": "40c9cdc9d5afd15f8c69bda48a0eb36d81df14adff5555de54a481df1baed5ca",
+          "sha256": "1a6b09fdf86ad0ad6aef167de8325670d6643c5d68c35e9846ff85270dcfed7a",
           "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/quality_gates.json",
           "url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/quality_gates.json"
         },
@@ -81,16 +81,16 @@
           "ok": true,
           "status_code": 200,
           "bytes": 8147,
-          "sha256": "40c9cdc9d5afd15f8c69bda48a0eb36d81df14adff5555de54a481df1baed5ca",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fquality_gates.json",
+          "sha256": "1a6b09fdf86ad0ad6aef167de8325670d6643c5d68c35e9846ff85270dcfed7a",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fquality_gates.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/quality_gates.json"
         },
         "hf_model": {
           "ok": true,
           "status_code": 200,
           "bytes": 8147,
-          "sha256": "40c9cdc9d5afd15f8c69bda48a0eb36d81df14adff5555de54a481df1baed5ca",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fquality_gates.json",
+          "sha256": "1a6b09fdf86ad0ad6aef167de8325670d6643c5d68c35e9846ff85270dcfed7a",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fquality_gates.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/quality_gates.json"
         }
       },
@@ -128,7 +128,7 @@
           "status_code": 200,
           "bytes": 4297465,
           "sha256": "89481cb9c11e923aef55dddd50b41a6fcc20ebf83efa94eb71df73b7782561f5",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fsingle_episode_explorer.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fsingle_episode_explorer.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/single_episode_explorer.json"
         },
         "hf_model": {
@@ -136,7 +136,7 @@
           "status_code": 200,
           "bytes": 4297465,
           "sha256": "89481cb9c11e923aef55dddd50b41a6fcc20ebf83efa94eb71df73b7782561f5",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fsingle_episode_explorer.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fsingle_episode_explorer.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/single_episode_explorer.json"
         }
       },
@@ -174,7 +174,7 @@
           "status_code": 200,
           "bytes": 2766828,
           "sha256": "19feb5e101ef50ec9f2a03e608ac75da68d642a2a38f20bd475635f0928279f0",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fsingle_episode_explorer.html",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fsingle_episode_explorer.html",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/single_episode_explorer.html"
         },
         "hf_model": {
@@ -182,7 +182,7 @@
           "status_code": 200,
           "bytes": 2766828,
           "sha256": "19feb5e101ef50ec9f2a03e608ac75da68d642a2a38f20bd475635f0928279f0",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/single_episode_explorer.html",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/single_episode_explorer.html",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/single_episode_explorer.html"
         }
       },
@@ -195,40 +195,40 @@
       "local": {
         "path": "docs/research_roadmap.html",
         "exists": true,
-        "bytes": 31320,
-        "sha256": "66f5c9fc7849c6a9d861710f76d9e7a1305e38951cf1dd86781c8c815a5c8a91"
+        "bytes": 31593,
+        "sha256": "05864ee10e582a4c05a5f1cdda4ae92a387c457375f4c7631fd303de96ad0c98"
       },
       "mirrors": {
         "github_pages": {
           "ok": true,
           "status_code": 200,
-          "bytes": 31320,
-          "sha256": "66f5c9fc7849c6a9d861710f76d9e7a1305e38951cf1dd86781c8c815a5c8a91",
+          "bytes": 31593,
+          "sha256": "05864ee10e582a4c05a5f1cdda4ae92a387c457375f4c7631fd303de96ad0c98",
           "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/research_roadmap.html",
           "url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/research_roadmap.html"
         },
         "hf_space": {
           "ok": true,
           "status_code": 200,
-          "bytes": 31320,
-          "sha256": "66f5c9fc7849c6a9d861710f76d9e7a1305e38951cf1dd86781c8c815a5c8a91",
+          "bytes": 31593,
+          "sha256": "05864ee10e582a4c05a5f1cdda4ae92a387c457375f4c7631fd303de96ad0c98",
           "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/research_roadmap.html",
           "url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/research_roadmap.html"
         },
         "hf_artifacts": {
           "ok": true,
           "status_code": 200,
-          "bytes": 31320,
-          "sha256": "66f5c9fc7849c6a9d861710f76d9e7a1305e38951cf1dd86781c8c815a5c8a91",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fresearch_roadmap.html",
+          "bytes": 31593,
+          "sha256": "05864ee10e582a4c05a5f1cdda4ae92a387c457375f4c7631fd303de96ad0c98",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fresearch_roadmap.html",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/research_roadmap.html"
         },
         "hf_model": {
           "ok": true,
           "status_code": 200,
-          "bytes": 31320,
-          "sha256": "66f5c9fc7849c6a9d861710f76d9e7a1305e38951cf1dd86781c8c815a5c8a91",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/research_roadmap.html",
+          "bytes": 31593,
+          "sha256": "05864ee10e582a4c05a5f1cdda4ae92a387c457375f4c7631fd303de96ad0c98",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/research_roadmap.html",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/research_roadmap.html"
         }
       },
@@ -266,7 +266,7 @@
           "status_code": 200,
           "bytes": 4072,
           "sha256": "7189be7fc773943859dc75a6119c1fb654d8905ea381e8fc9f782d4b33f5e466",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/results%2Fsingle_episode_diagnostics%2Fprovenance.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/results%2Fsingle_episode_diagnostics%2Fprovenance.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/results/single_episode_diagnostics/provenance.json"
         },
         "hf_model": {
@@ -274,7 +274,7 @@
           "status_code": 200,
           "bytes": 4072,
           "sha256": "7189be7fc773943859dc75a6119c1fb654d8905ea381e8fc9f782d4b33f5e466",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/results%2Fsingle_episode_diagnostics%2Fprovenance.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/results%2Fsingle_episode_diagnostics%2Fprovenance.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/results/single_episode_diagnostics/provenance.json"
         }
       },
@@ -312,7 +312,7 @@
           "status_code": 200,
           "bytes": 995,
           "sha256": "813320bbe5d57c74803f5dfb080d440ef1804af41950e031d82ab80311eb04c7",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/results%2Fsingle_episode_diagnostics%2Fobject_labels%2Fobject_vocab.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/results%2Fsingle_episode_diagnostics%2Fobject_labels%2Fobject_vocab.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/results/single_episode_diagnostics/object_labels/object_vocab.json"
         },
         "hf_model": {
@@ -320,7 +320,7 @@
           "status_code": 200,
           "bytes": 995,
           "sha256": "813320bbe5d57c74803f5dfb080d440ef1804af41950e031d82ab80311eb04c7",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/results%2Fsingle_episode_diagnostics%2Fobject_labels%2Fobject_vocab.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/results%2Fsingle_episode_diagnostics%2Fobject_labels%2Fobject_vocab.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/results/single_episode_diagnostics/object_labels/object_vocab.json"
         }
       },
@@ -333,40 +333,40 @@
       "local": {
         "path": "docs/data/public_surface_qa.json",
         "exists": true,
-        "bytes": 5647,
-        "sha256": "46576b574f962767b3d52ffd7075659d72ddf3d11154e47997fe7564de1731cb"
+        "bytes": 5631,
+        "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624"
       },
       "mirrors": {
         "github_pages": {
           "ok": true,
           "status_code": 200,
-          "bytes": 5647,
-          "sha256": "46576b574f962767b3d52ffd7075659d72ddf3d11154e47997fe7564de1731cb",
+          "bytes": 5631,
+          "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624",
           "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/public_surface_qa.json",
           "url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/public_surface_qa.json"
         },
         "hf_space": {
           "ok": true,
           "status_code": 200,
-          "bytes": 5647,
-          "sha256": "46576b574f962767b3d52ffd7075659d72ddf3d11154e47997fe7564de1731cb",
+          "bytes": 5631,
+          "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624",
           "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/public_surface_qa.json",
           "url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/public_surface_qa.json"
         },
         "hf_artifacts": {
           "ok": true,
           "status_code": 200,
-          "bytes": 5647,
-          "sha256": "46576b574f962767b3d52ffd7075659d72ddf3d11154e47997fe7564de1731cb",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fpublic_surface_qa.json",
+          "bytes": 5631,
+          "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fpublic_surface_qa.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/public_surface_qa.json"
         },
         "hf_model": {
           "ok": true,
           "status_code": 200,
-          "bytes": 5647,
-          "sha256": "46576b574f962767b3d52ffd7075659d72ddf3d11154e47997fe7564de1731cb",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fpublic_surface_qa.json",
+          "bytes": 5631,
+          "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fpublic_surface_qa.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/public_surface_qa.json"
         }
       },
@@ -404,7 +404,7 @@
           "status_code": 200,
           "bytes": 4032,
           "sha256": "2390c3663e9a110b9ed2f9c581b88a95bbd811a72ffb3e8dc780dae420e0f919",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Frendered_site_check.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Frendered_site_check.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/rendered_site_check.json"
         },
         "hf_model": {
@@ -412,7 +412,7 @@
           "status_code": 200,
           "bytes": 4032,
           "sha256": "2390c3663e9a110b9ed2f9c581b88a95bbd811a72ffb3e8dc780dae420e0f919",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Frendered_site_check.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Frendered_site_check.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/rendered_site_check.json"
         }
       },
@@ -425,40 +425,40 @@
       "local": {
         "path": "docs/data/xperience10m_dataset_card_alignment.json",
         "exists": true,
-        "bytes": 7573,
-        "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7"
+        "bytes": 7585,
+        "sha256": "8bb8f22cf1cc8938f7dcfef15d11e3e544243bb56dc32c792313d224e91a8a2d"
       },
       "mirrors": {
         "github_pages": {
           "ok": true,
           "status_code": 200,
-          "bytes": 7573,
-          "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7",
+          "bytes": 7585,
+          "sha256": "8bb8f22cf1cc8938f7dcfef15d11e3e544243bb56dc32c792313d224e91a8a2d",
           "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/xperience10m_dataset_card_alignment.json",
           "url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/xperience10m_dataset_card_alignment.json"
         },
         "hf_space": {
           "ok": true,
           "status_code": 200,
-          "bytes": 7573,
-          "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7",
+          "bytes": 7585,
+          "sha256": "8bb8f22cf1cc8938f7dcfef15d11e3e544243bb56dc32c792313d224e91a8a2d",
           "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/xperience10m_dataset_card_alignment.json",
           "url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/xperience10m_dataset_card_alignment.json"
         },
         "hf_artifacts": {
           "ok": true,
           "status_code": 200,
-          "bytes": 7573,
-          "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fxperience10m_dataset_card_alignment.json",
+          "bytes": 7585,
+          "sha256": "8bb8f22cf1cc8938f7dcfef15d11e3e544243bb56dc32c792313d224e91a8a2d",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fxperience10m_dataset_card_alignment.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/xperience10m_dataset_card_alignment.json"
         },
         "hf_model": {
           "ok": true,
           "status_code": 200,
-          "bytes": 7573,
-          "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fxperience10m_dataset_card_alignment.json",
+          "bytes": 7585,
+          "sha256": "8bb8f22cf1cc8938f7dcfef15d11e3e544243bb56dc32c792313d224e91a8a2d",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fxperience10m_dataset_card_alignment.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/xperience10m_dataset_card_alignment.json"
         }
       },
@@ -472,14 +472,14 @@
         "path": "docs/data/source_alignment_audit.json",
         "exists": true,
         "bytes": 4432,
-        "sha256": "9d8eb05aa29f898a28f64b0b23ef0b912ba0f43d6ef558d213e05619da0b3197"
+        "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
       },
       "mirrors": {
         "github_pages": {
           "ok": true,
           "status_code": 200,
           "bytes": 4432,
-          "sha256": "9d8eb05aa29f898a28f64b0b23ef0b912ba0f43d6ef558d213e05619da0b3197",
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b",
           "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/source_alignment_audit.json",
           "url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/source_alignment_audit.json"
         },
@@ -487,7 +487,7 @@
           "ok": true,
           "status_code": 200,
           "bytes": 4432,
-          "sha256": "9d8eb05aa29f898a28f64b0b23ef0b912ba0f43d6ef558d213e05619da0b3197",
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b",
           "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/source_alignment_audit.json",
           "url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/source_alignment_audit.json"
         },
@@ -495,16 +495,16 @@
           "ok": true,
           "status_code": 200,
           "bytes": 4432,
-          "sha256": "9d8eb05aa29f898a28f64b0b23ef0b912ba0f43d6ef558d213e05619da0b3197",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fsource_alignment_audit.json",
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fsource_alignment_audit.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/source_alignment_audit.json"
         },
         "hf_model": {
           "ok": true,
           "status_code": 200,
           "bytes": 4432,
-          "sha256": "9d8eb05aa29f898a28f64b0b23ef0b912ba0f43d6ef558d213e05619da0b3197",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fsource_alignment_audit.json",
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fsource_alignment_audit.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/source_alignment_audit.json"
         }
       },
@@ -517,40 +517,40 @@
       "local": {
         "path": "docs/data/project_status.json",
         "exists": true,
-        "bytes": 7340,
-        "sha256": "3f54a794582f4b61f60fe240a9ea6f1782e29a21498e2d63dad670495d4e0d39"
+        "bytes": 9006,
+        "sha256": "38d289ce14e7b4f0998ce156718662f2298ee3aa4e77782a40316ad33aefaefb"
       },
       "mirrors": {
         "github_pages": {
           "ok": true,
           "status_code": 200,
-          "bytes": 7340,
-          "sha256": "3f54a794582f4b61f60fe240a9ea6f1782e29a21498e2d63dad670495d4e0d39",
+          "bytes": 9006,
+          "sha256": "38d289ce14e7b4f0998ce156718662f2298ee3aa4e77782a40316ad33aefaefb",
           "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/project_status.json",
           "url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/project_status.json"
         },
         "hf_space": {
           "ok": true,
           "status_code": 200,
-          "bytes": 7340,
-          "sha256": "3f54a794582f4b61f60fe240a9ea6f1782e29a21498e2d63dad670495d4e0d39",
+          "bytes": 9006,
+          "sha256": "38d289ce14e7b4f0998ce156718662f2298ee3aa4e77782a40316ad33aefaefb",
           "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/project_status.json",
           "url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/project_status.json"
         },
         "hf_artifacts": {
           "ok": true,
           "status_code": 200,
-          "bytes": 7340,
-          "sha256": "3f54a794582f4b61f60fe240a9ea6f1782e29a21498e2d63dad670495d4e0d39",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fproject_status.json",
+          "bytes": 9006,
+          "sha256": "38d289ce14e7b4f0998ce156718662f2298ee3aa4e77782a40316ad33aefaefb",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fproject_status.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/project_status.json"
         },
         "hf_model": {
           "ok": true,
           "status_code": 200,
-          "bytes": 7340,
-          "sha256": "3f54a794582f4b61f60fe240a9ea6f1782e29a21498e2d63dad670495d4e0d39",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fproject_status.json",
+          "bytes": 9006,
+          "sha256": "38d289ce14e7b4f0998ce156718662f2298ee3aa4e77782a40316ad33aefaefb",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fproject_status.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/project_status.json"
         }
       },
@@ -588,7 +588,7 @@
           "status_code": 200,
           "bytes": 13644,
           "sha256": "69281b1727d3de7a734fe89846e9881b02fa1df4108f4ae39946acbcdf6ed8bf",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fevaluation_protocol.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fevaluation_protocol.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/evaluation_protocol.json"
         },
         "hf_model": {
@@ -596,7 +596,7 @@
           "status_code": 200,
           "bytes": 13644,
           "sha256": "69281b1727d3de7a734fe89846e9881b02fa1df4108f4ae39946acbcdf6ed8bf",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fevaluation_protocol.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fevaluation_protocol.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/evaluation_protocol.json"
         }
       },
@@ -609,40 +609,40 @@
       "local": {
         "path": "docs/data/research_takeaways.json",
         "exists": true,
-        "bytes": 5356,
-        "sha256": "0621f28d602582c70524cd9455b01558e7003267b1f4786d005e31597e46b051"
+        "bytes": 6814,
+        "sha256": "538ec80950dbf5d77cc6ab2bd5a96a3af191b56a257b895e48b6cf28deb0d044"
       },
       "mirrors": {
         "github_pages": {
           "ok": true,
           "status_code": 200,
-          "bytes": 5356,
-          "sha256": "0621f28d602582c70524cd9455b01558e7003267b1f4786d005e31597e46b051",
+          "bytes": 6814,
+          "sha256": "538ec80950dbf5d77cc6ab2bd5a96a3af191b56a257b895e48b6cf28deb0d044",
           "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/research_takeaways.json",
           "url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/research_takeaways.json"
         },
         "hf_space": {
           "ok": true,
           "status_code": 200,
-          "bytes": 5356,
-          "sha256": "0621f28d602582c70524cd9455b01558e7003267b1f4786d005e31597e46b051",
+          "bytes": 6814,
+          "sha256": "538ec80950dbf5d77cc6ab2bd5a96a3af191b56a257b895e48b6cf28deb0d044",
           "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/research_takeaways.json",
           "url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/research_takeaways.json"
         },
         "hf_artifacts": {
           "ok": true,
           "status_code": 200,
-          "bytes": 5356,
-          "sha256": "0621f28d602582c70524cd9455b01558e7003267b1f4786d005e31597e46b051",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fresearch_takeaways.json",
+          "bytes": 6814,
+          "sha256": "538ec80950dbf5d77cc6ab2bd5a96a3af191b56a257b895e48b6cf28deb0d044",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fresearch_takeaways.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/research_takeaways.json"
         },
         "hf_model": {
           "ok": true,
           "status_code": 200,
-          "bytes": 5356,
-          "sha256": "0621f28d602582c70524cd9455b01558e7003267b1f4786d005e31597e46b051",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fresearch_takeaways.json",
+          "bytes": 6814,
+          "sha256": "538ec80950dbf5d77cc6ab2bd5a96a3af191b56a257b895e48b6cf28deb0d044",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fresearch_takeaways.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/research_takeaways.json"
         }
       },
@@ -655,40 +655,40 @@
       "local": {
         "path": "docs/data/research_roadmap.json",
         "exists": true,
-        "bytes": 4594,
-        "sha256": "bbce65e803a0fad55cc739eb2dc6fdf50a69c81362f05d216324d0813ee4ccad"
+        "bytes": 5752,
+        "sha256": "0510e919ad9ffa3aebe052273e90133a57572c3b21b991db1ff28de83061f6ca"
       },
       "mirrors": {
         "github_pages": {
           "ok": true,
           "status_code": 200,
-          "bytes": 4594,
-          "sha256": "bbce65e803a0fad55cc739eb2dc6fdf50a69c81362f05d216324d0813ee4ccad",
+          "bytes": 5752,
+          "sha256": "0510e919ad9ffa3aebe052273e90133a57572c3b21b991db1ff28de83061f6ca",
           "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/research_roadmap.json",
           "url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/research_roadmap.json"
         },
         "hf_space": {
           "ok": true,
           "status_code": 200,
-          "bytes": 4594,
-          "sha256": "bbce65e803a0fad55cc739eb2dc6fdf50a69c81362f05d216324d0813ee4ccad",
+          "bytes": 5752,
+          "sha256": "0510e919ad9ffa3aebe052273e90133a57572c3b21b991db1ff28de83061f6ca",
           "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/research_roadmap.json",
           "url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/research_roadmap.json"
         },
         "hf_artifacts": {
           "ok": true,
           "status_code": 200,
-          "bytes": 4594,
-          "sha256": "bbce65e803a0fad55cc739eb2dc6fdf50a69c81362f05d216324d0813ee4ccad",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fresearch_roadmap.json",
+          "bytes": 5752,
+          "sha256": "0510e919ad9ffa3aebe052273e90133a57572c3b21b991db1ff28de83061f6ca",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fresearch_roadmap.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/research_roadmap.json"
         },
         "hf_model": {
           "ok": true,
           "status_code": 200,
-          "bytes": 4594,
-          "sha256": "bbce65e803a0fad55cc739eb2dc6fdf50a69c81362f05d216324d0813ee4ccad",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fresearch_roadmap.json",
+          "bytes": 5752,
+          "sha256": "0510e919ad9ffa3aebe052273e90133a57572c3b21b991db1ff28de83061f6ca",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fresearch_roadmap.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/research_roadmap.json"
         }
       },
@@ -701,40 +701,40 @@
       "local": {
         "path": "docs/data/research_roadmap_interactive.json",
         "exists": true,
-        "bytes": 121327,
-        "sha256": "bd3626b9217260437120304e0509d95f75de491e76a86ad284790109763e04b0"
+        "bytes": 131526,
+        "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0"
       },
       "mirrors": {
         "github_pages": {
           "ok": true,
           "status_code": 200,
-          "bytes": 121327,
-          "sha256": "bd3626b9217260437120304e0509d95f75de491e76a86ad284790109763e04b0",
+          "bytes": 131526,
+          "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0",
           "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/research_roadmap_interactive.json",
           "url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/research_roadmap_interactive.json"
         },
         "hf_space": {
           "ok": true,
           "status_code": 200,
-          "bytes": 121327,
-          "sha256": "bd3626b9217260437120304e0509d95f75de491e76a86ad284790109763e04b0",
+          "bytes": 131526,
+          "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0",
           "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/research_roadmap_interactive.json",
           "url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/research_roadmap_interactive.json"
         },
         "hf_artifacts": {
           "ok": true,
           "status_code": 200,
-          "bytes": 121327,
-          "sha256": "bd3626b9217260437120304e0509d95f75de491e76a86ad284790109763e04b0",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fresearch_roadmap_interactive.json",
+          "bytes": 131526,
+          "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fresearch_roadmap_interactive.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/research_roadmap_interactive.json"
         },
         "hf_model": {
           "ok": true,
           "status_code": 200,
-          "bytes": 121327,
-          "sha256": "bd3626b9217260437120304e0509d95f75de491e76a86ad284790109763e04b0",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fresearch_roadmap_interactive.json",
+          "bytes": 131526,
+          "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fresearch_roadmap_interactive.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/research_roadmap_interactive.json"
         }
       },
@@ -748,14 +748,14 @@
         "path": "docs/data/figure_index.json",
         "exists": true,
         "bytes": 13434,
-        "sha256": "fc7c9d504509bea2bab08238c96b745f00fdd6a016019d121d4bf0e22b5f3936"
+        "sha256": "1a28548d2838b5a5ea9464761ccb73a3e55d6d48dd4c68d321f056bfdc4ad560"
       },
       "mirrors": {
         "github_pages": {
           "ok": true,
           "status_code": 200,
           "bytes": 13434,
-          "sha256": "fc7c9d504509bea2bab08238c96b745f00fdd6a016019d121d4bf0e22b5f3936",
+          "sha256": "1a28548d2838b5a5ea9464761ccb73a3e55d6d48dd4c68d321f056bfdc4ad560",
           "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/figure_index.json",
           "url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/figure_index.json"
         },
@@ -763,7 +763,7 @@
           "ok": true,
           "status_code": 200,
           "bytes": 13434,
-          "sha256": "fc7c9d504509bea2bab08238c96b745f00fdd6a016019d121d4bf0e22b5f3936",
+          "sha256": "1a28548d2838b5a5ea9464761ccb73a3e55d6d48dd4c68d321f056bfdc4ad560",
           "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/figure_index.json",
           "url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/figure_index.json"
         },
@@ -771,16 +771,16 @@
           "ok": true,
           "status_code": 200,
           "bytes": 13434,
-          "sha256": "fc7c9d504509bea2bab08238c96b745f00fdd6a016019d121d4bf0e22b5f3936",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Ffigure_index.json",
+          "sha256": "1a28548d2838b5a5ea9464761ccb73a3e55d6d48dd4c68d321f056bfdc4ad560",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Ffigure_index.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/figure_index.json"
         },
         "hf_model": {
           "ok": true,
           "status_code": 200,
           "bytes": 13434,
-          "sha256": "fc7c9d504509bea2bab08238c96b745f00fdd6a016019d121d4bf0e22b5f3936",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Ffigure_index.json",
+          "sha256": "1a28548d2838b5a5ea9464761ccb73a3e55d6d48dd4c68d321f056bfdc4ad560",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Ffigure_index.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/figure_index.json"
         }
       },
@@ -818,7 +818,7 @@
           "status_code": 200,
           "bytes": 26976,
           "sha256": "4a43c4df17f26cd02c988afab34165a9b106acf440cf5a09d9f042b6e1fada1b",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Ftask_walkthroughs.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Ftask_walkthroughs.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/task_walkthroughs.json"
         },
         "hf_model": {
@@ -826,7 +826,7 @@
           "status_code": 200,
           "bytes": 26976,
           "sha256": "4a43c4df17f26cd02c988afab34165a9b106acf440cf5a09d9f042b6e1fada1b",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Ftask_walkthroughs.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Ftask_walkthroughs.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/task_walkthroughs.json"
         }
       },
@@ -840,14 +840,14 @@
         "path": "docs/data/task_surface_integrity.json",
         "exists": true,
         "bytes": 45779,
-        "sha256": "8dfb556fc81d931bcaf547c52442e39c5f0579c8cc6508afa117c15aa879e8cb"
+        "sha256": "ea0a9de1e414e275b41eca27908c2d5575da3c05e05d7f2544122d763bbdfd33"
       },
       "mirrors": {
         "github_pages": {
           "ok": true,
           "status_code": 200,
           "bytes": 45779,
-          "sha256": "8dfb556fc81d931bcaf547c52442e39c5f0579c8cc6508afa117c15aa879e8cb",
+          "sha256": "ea0a9de1e414e275b41eca27908c2d5575da3c05e05d7f2544122d763bbdfd33",
           "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/task_surface_integrity.json",
           "url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/task_surface_integrity.json"
         },
@@ -855,7 +855,7 @@
           "ok": true,
           "status_code": 200,
           "bytes": 45779,
-          "sha256": "8dfb556fc81d931bcaf547c52442e39c5f0579c8cc6508afa117c15aa879e8cb",
+          "sha256": "ea0a9de1e414e275b41eca27908c2d5575da3c05e05d7f2544122d763bbdfd33",
           "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/task_surface_integrity.json",
           "url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/task_surface_integrity.json"
         },
@@ -863,16 +863,16 @@
           "ok": true,
           "status_code": 200,
           "bytes": 45779,
-          "sha256": "8dfb556fc81d931bcaf547c52442e39c5f0579c8cc6508afa117c15aa879e8cb",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Ftask_surface_integrity.json",
+          "sha256": "ea0a9de1e414e275b41eca27908c2d5575da3c05e05d7f2544122d763bbdfd33",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Ftask_surface_integrity.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/task_surface_integrity.json"
         },
         "hf_model": {
           "ok": true,
           "status_code": 200,
           "bytes": 45779,
-          "sha256": "8dfb556fc81d931bcaf547c52442e39c5f0579c8cc6508afa117c15aa879e8cb",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Ftask_surface_integrity.json",
+          "sha256": "ea0a9de1e414e275b41eca27908c2d5575da3c05e05d7f2544122d763bbdfd33",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Ftask_surface_integrity.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/task_surface_integrity.json"
         }
       },
@@ -910,7 +910,7 @@
           "status_code": 200,
           "bytes": 3897,
           "sha256": "0d4e77e3e5903e1d6eff69cc487ad2b7f8d89ad06a44de09d0edca47e878b7d2",
-          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fbrand_assets.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fbrand_assets.json",
           "url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/resolve/main/docs/data/brand_assets.json"
         },
         "hf_model": {
@@ -918,7 +918,7 @@
           "status_code": 200,
           "bytes": 3897,
           "sha256": "0d4e77e3e5903e1d6eff69cc487ad2b7f8d89ad06a44de09d0edca47e878b7d2",
-          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fbrand_assets.json",
+          "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fbrand_assets.json",
           "url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/resolve/main/metrics/brand_assets.json"
         }
       },
@@ -1026,8 +1026,8 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 166238,
-        "sha256": "4ccdf2d1867793cfa64ff7475ff091300e135935140bd65b3277ac999a317747",
+        "bytes": 170464,
+        "sha256": "f85f66d3551bf641c0f21a2a7581c3cc9df277c3f22403c8a361346a18c11385",
         "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/"
       },
       "required_marker_count": 23,
@@ -1043,8 +1043,8 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 166238,
-        "sha256": "4ccdf2d1867793cfa64ff7475ff091300e135935140bd65b3277ac999a317747",
+        "bytes": 170464,
+        "sha256": "f85f66d3551bf641c0f21a2a7581c3cc9df277c3f22403c8a361346a18c11385",
         "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/index.html"
       },
       "required_marker_count": 23,
@@ -1060,8 +1060,8 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 3692,
-        "sha256": "521619031c044f2088e14068bb1843bb3480a1d2367451b55acbdc0d09af3819",
+        "bytes": 4594,
+        "sha256": "7c388415f9407a75e9bad4a53b8a1a675328356c0f058b64e00887c28914ca5c",
         "final_url": "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/raw/main/README.md"
       },
       "required_marker_count": 17,
@@ -1111,8 +1111,8 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 9107,
-        "sha256": "329f6b835933899df04822ee08433970ab557ff5c6c7a5541d35d53841ac11b8",
+        "bytes": 63788,
+        "sha256": "01eab741ef14e1b4d03112b6ca9554eeec70b183efe012008c2531d0010073cb",
         "final_url": "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines/raw/main/README.md"
       },
       "required_marker_count": 17,
@@ -1128,8 +1128,8 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 97337,
-        "sha256": "ba4a25793a4204952d5e8aa770906965e005f55716b09b292b538606bec18da2",
+        "bytes": 108606,
+        "sha256": "35a4d0a5f957fabb49cd46c859f389ba9f333294d0076f69bcb80f0da6b08d24",
         "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/mirror_parity.json"
       },
       "required_marker_count": 1,
@@ -1145,8 +1145,8 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 97337,
-        "sha256": "ba4a25793a4204952d5e8aa770906965e005f55716b09b292b538606bec18da2",
+        "bytes": 108606,
+        "sha256": "35a4d0a5f957fabb49cd46c859f389ba9f333294d0076f69bcb80f0da6b08d24",
         "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/mirror_parity.json"
       },
       "required_marker_count": 1,
@@ -1162,9 +1162,9 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 97337,
-        "sha256": "ba4a25793a4204952d5e8aa770906965e005f55716b09b292b538606bec18da2",
-        "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fmirror_parity.json"
+        "bytes": 108606,
+        "sha256": "35a4d0a5f957fabb49cd46c859f389ba9f333294d0076f69bcb80f0da6b08d24",
+        "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fmirror_parity.json"
       },
       "required_marker_count": 1,
       "missing_markers": [],
@@ -1179,9 +1179,9 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 97337,
-        "sha256": "ba4a25793a4204952d5e8aa770906965e005f55716b09b292b538606bec18da2",
-        "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fmirror_parity.json"
+        "bytes": 108606,
+        "sha256": "35a4d0a5f957fabb49cd46c859f389ba9f333294d0076f69bcb80f0da6b08d24",
+        "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fmirror_parity.json"
       },
       "required_marker_count": 1,
       "missing_markers": [],
@@ -1197,7 +1197,7 @@
         "ok": true,
         "status_code": 200,
         "bytes": 7289,
-        "sha256": "b2292107e5472a74e5b78dc769d2ff2739c41d7773367d9868b9d57db99a77e1",
+        "sha256": "8bd9ff61443b39b48829f80d7d4ff2193a37402e22cf842023c469c3584480af",
         "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/publication_audit.json"
       },
       "required_marker_count": 1,
@@ -1214,7 +1214,7 @@
         "ok": true,
         "status_code": 200,
         "bytes": 7289,
-        "sha256": "b2292107e5472a74e5b78dc769d2ff2739c41d7773367d9868b9d57db99a77e1",
+        "sha256": "8bd9ff61443b39b48829f80d7d4ff2193a37402e22cf842023c469c3584480af",
         "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/publication_audit.json"
       },
       "required_marker_count": 1,
@@ -1231,8 +1231,8 @@
         "ok": true,
         "status_code": 200,
         "bytes": 7289,
-        "sha256": "b2292107e5472a74e5b78dc769d2ff2739c41d7773367d9868b9d57db99a77e1",
-        "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fpublication_audit.json"
+        "sha256": "8bd9ff61443b39b48829f80d7d4ff2193a37402e22cf842023c469c3584480af",
+        "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fpublication_audit.json"
       },
       "required_marker_count": 1,
       "missing_markers": [],
@@ -1248,8 +1248,8 @@
         "ok": true,
         "status_code": 200,
         "bytes": 7289,
-        "sha256": "b2292107e5472a74e5b78dc769d2ff2739c41d7773367d9868b9d57db99a77e1",
-        "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fpublication_audit.json"
+        "sha256": "8bd9ff61443b39b48829f80d7d4ff2193a37402e22cf842023c469c3584480af",
+        "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fpublication_audit.json"
       },
       "required_marker_count": 1,
       "missing_markers": [],
@@ -1264,8 +1264,8 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 14312,
-        "sha256": "5c86a8f27a05c1b0d9f8a407b73be2db1fbd88aafc64687f9f07b9bc4f397305",
+        "bytes": 14718,
+        "sha256": "3cb852491d4b7998cd1c2b04a644c9d97eaddf4b8bd29bb00f971ed2e880f4c5",
         "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/website_integrity.json"
       },
       "required_marker_count": 1,
@@ -1281,8 +1281,8 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 14312,
-        "sha256": "5c86a8f27a05c1b0d9f8a407b73be2db1fbd88aafc64687f9f07b9bc4f397305",
+        "bytes": 14718,
+        "sha256": "3cb852491d4b7998cd1c2b04a644c9d97eaddf4b8bd29bb00f971ed2e880f4c5",
         "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/website_integrity.json"
       },
       "required_marker_count": 1,
@@ -1298,9 +1298,9 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 14312,
-        "sha256": "5c86a8f27a05c1b0d9f8a407b73be2db1fbd88aafc64687f9f07b9bc4f397305",
-        "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fwebsite_integrity.json"
+        "bytes": 14718,
+        "sha256": "3cb852491d4b7998cd1c2b04a644c9d97eaddf4b8bd29bb00f971ed2e880f4c5",
+        "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fwebsite_integrity.json"
       },
       "required_marker_count": 1,
       "missing_markers": [],
@@ -1315,9 +1315,9 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 14312,
-        "sha256": "5c86a8f27a05c1b0d9f8a407b73be2db1fbd88aafc64687f9f07b9bc4f397305",
-        "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fwebsite_integrity.json"
+        "bytes": 14718,
+        "sha256": "3cb852491d4b7998cd1c2b04a644c9d97eaddf4b8bd29bb00f971ed2e880f4c5",
+        "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fwebsite_integrity.json"
       },
       "required_marker_count": 1,
       "missing_markers": [],
@@ -1332,8 +1332,8 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 5647,
-        "sha256": "46576b574f962767b3d52ffd7075659d72ddf3d11154e47997fe7564de1731cb",
+        "bytes": 5631,
+        "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624",
         "final_url": "https://chaoyue0307.github.io/ropedia-xperience-10m-task-suite/data/public_surface_qa.json"
       },
       "required_marker_count": 1,
@@ -1349,8 +1349,8 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 5647,
-        "sha256": "46576b574f962767b3d52ffd7075659d72ddf3d11154e47997fe7564de1731cb",
+        "bytes": 5631,
+        "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624",
         "final_url": "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite/raw/main/data/public_surface_qa.json"
       },
       "required_marker_count": 1,
@@ -1366,9 +1366,9 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 5647,
-        "sha256": "46576b574f962767b3d52ffd7075659d72ddf3d11154e47997fe7564de1731cb",
-        "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/91c355fe10e26e62be8d7dad2b332a7264287a1d/docs%2Fdata%2Fpublic_surface_qa.json"
+        "bytes": 5631,
+        "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624",
+        "final_url": "https://huggingface.co/api/resolve-cache/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts/e099e743cb44004fbcd780eb3b049fd2c85063ae/docs%2Fdata%2Fpublic_surface_qa.json"
       },
       "required_marker_count": 1,
       "missing_markers": [],
@@ -1383,9 +1383,9 @@
       "fetch": {
         "ok": true,
         "status_code": 200,
-        "bytes": 5647,
-        "sha256": "46576b574f962767b3d52ffd7075659d72ddf3d11154e47997fe7564de1731cb",
-        "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/0816a5a322b7f4ec0b26d6308a4a4cfd4a3de141/metrics%2Fpublic_surface_qa.json"
+        "bytes": 5631,
+        "sha256": "04327d22b4210afef123a29071ec7d686854efac6ca6c1b301b58fa9a04a2624",
+        "final_url": "https://huggingface.co/api/resolve-cache/models/cy0307/ropedia-xperience-10m-task-baselines/e0ca7d8c493d20f2379b2372f10b0317512ca371/metrics%2Fpublic_surface_qa.json"
       },
       "required_marker_count": 1,
       "missing_markers": [],

docs/data/mirror_parity.json

@@ -1,20 +1,16 @@
 {
-  "status": "fail",
-  "generated_at_utc": "2026-06-03T16:58:21+00:00",
+  "status": "pass",
+  "generated_at_utc": "2026-06-03T19:30:53+00:00",
   "hf_root": "hf_publish",
   "summary": {
     "group_count": 101,
-    "failure_count": 6,
-    "failures_by_surface": {
-      "hf_space": 2,
-      "hf_artifacts": 2,
-      "hf_model": 2
-    }
+    "failure_count": 0,
+    "failures_by_surface": {}
   },
   "checks": [
     {
       "name": "repo_hf_space_artifact_model_data_parity",
-      "status": "fail"
+      "status": "pass"
     },
     {
       "name": "repo_hf_visual_asset_parity",
@@ -34,7 +30,7 @@
     },
     {
       "name": "repo_hf_quality_doc_parity",
-      "status": "fail"
+      "status": "pass"
     }
   ],
   "groups": [
@@ -44,27 +40,27 @@
       "local": {
         "path": "repo:docs/data/audio_ablation_summary.json",
         "exists": true,
-        "bytes": 9735,
-        "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
+        "bytes": 9701,
+        "sha256": "7debd9e67b7df09322d743193bf8b785277a992c9e8c82c08c7a36c4e066e6de"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/audio_ablation_summary.json",
           "exists": true,
-          "bytes": 9735,
-          "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
+          "bytes": 9701,
+          "sha256": "7debd9e67b7df09322d743193bf8b785277a992c9e8c82c08c7a36c4e066e6de"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/audio_ablation_summary.json",
           "exists": true,
-          "bytes": 9735,
-          "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
+          "bytes": 9701,
+          "sha256": "7debd9e67b7df09322d743193bf8b785277a992c9e8c82c08c7a36c4e066e6de"
         },
         "hf_model": {
           "path": "hf_model:metrics/audio_ablation_summary.json",
           "exists": true,
-          "bytes": 9735,
-          "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
+          "bytes": 9701,
+          "sha256": "7debd9e67b7df09322d743193bf8b785277a992c9e8c82c08c7a36c4e066e6de"
         }
       },
       "failures": []
@@ -75,27 +71,27 @@
       "local": {
         "path": "repo:docs/data/artifact_index.json",
         "exists": true,
-        "bytes": 32378,
-        "sha256": "553085b7e13b5d4ba9875bd7f59b376cd37094533e20094a54712bd46ed01ced"
+        "bytes": 32292,
+        "sha256": "dcd253ba6636ea4bf78260b51a1493b4faf7cf5fa22e74d67d4cd277b9fe4189"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/artifact_index.json",
           "exists": true,
-          "bytes": 32378,
-          "sha256": "553085b7e13b5d4ba9875bd7f59b376cd37094533e20094a54712bd46ed01ced"
+          "bytes": 32292,
+          "sha256": "dcd253ba6636ea4bf78260b51a1493b4faf7cf5fa22e74d67d4cd277b9fe4189"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/artifact_index.json",
           "exists": true,
-          "bytes": 32378,
-          "sha256": "553085b7e13b5d4ba9875bd7f59b376cd37094533e20094a54712bd46ed01ced"
+          "bytes": 32292,
+          "sha256": "dcd253ba6636ea4bf78260b51a1493b4faf7cf5fa22e74d67d4cd277b9fe4189"
         },
         "hf_model": {
           "path": "hf_model:metrics/artifact_index.json",
           "exists": true,
-          "bytes": 32378,
-          "sha256": "553085b7e13b5d4ba9875bd7f59b376cd37094533e20094a54712bd46ed01ced"
+          "bytes": 32292,
+          "sha256": "dcd253ba6636ea4bf78260b51a1493b4faf7cf5fa22e74d67d4cd277b9fe4189"
         }
       },
       "failures": []
@@ -137,27 +133,27 @@
       "local": {
         "path": "repo:docs/data/evidence_contract.json",
         "exists": true,
-        "bytes": 12025,
-        "sha256": "a69ae4f0c97d1c8491d5452fe4594039d7b882f56e5d287208a90d2bb9f80c00"
+        "bytes": 12096,
+        "sha256": "3d2815d31952a5102c78818a852db8ff36faf49d89643b32ad59bcda91d499cb"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/evidence_contract.json",
           "exists": true,
-          "bytes": 12025,
-          "sha256": "a69ae4f0c97d1c8491d5452fe4594039d7b882f56e5d287208a90d2bb9f80c00"
+          "bytes": 12096,
+          "sha256": "3d2815d31952a5102c78818a852db8ff36faf49d89643b32ad59bcda91d499cb"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/evidence_contract.json",
           "exists": true,
-          "bytes": 12025,
-          "sha256": "a69ae4f0c97d1c8491d5452fe4594039d7b882f56e5d287208a90d2bb9f80c00"
+          "bytes": 12096,
+          "sha256": "3d2815d31952a5102c78818a852db8ff36faf49d89643b32ad59bcda91d499cb"
         },
         "hf_model": {
           "path": "hf_model:metrics/evidence_contract.json",
           "exists": true,
-          "bytes": 12025,
-          "sha256": "a69ae4f0c97d1c8491d5452fe4594039d7b882f56e5d287208a90d2bb9f80c00"
+          "bytes": 12096,
+          "sha256": "3d2815d31952a5102c78818a852db8ff36faf49d89643b32ad59bcda91d499cb"
         }
       },
       "failures": []
@@ -168,27 +164,27 @@
       "local": {
         "path": "repo:docs/data/evaluation_protocol.json",
         "exists": true,
-        "bytes": 13644,
-        "sha256": "69281b1727d3de7a734fe89846e9881b02fa1df4108f4ae39946acbcdf6ed8bf"
+        "bytes": 13686,
+        "sha256": "a2a35bd3b05e228c99c5ef7cc002c84708dcaff893cd01d51b011dadbf2c02c4"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/evaluation_protocol.json",
           "exists": true,
-          "bytes": 13644,
-          "sha256": "69281b1727d3de7a734fe89846e9881b02fa1df4108f4ae39946acbcdf6ed8bf"
+          "bytes": 13686,
+          "sha256": "a2a35bd3b05e228c99c5ef7cc002c84708dcaff893cd01d51b011dadbf2c02c4"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/evaluation_protocol.json",
           "exists": true,
-          "bytes": 13644,
-          "sha256": "69281b1727d3de7a734fe89846e9881b02fa1df4108f4ae39946acbcdf6ed8bf"
+          "bytes": 13686,
+          "sha256": "a2a35bd3b05e228c99c5ef7cc002c84708dcaff893cd01d51b011dadbf2c02c4"
         },
         "hf_model": {
           "path": "hf_model:metrics/evaluation_protocol.json",
           "exists": true,
-          "bytes": 13644,
-          "sha256": "69281b1727d3de7a734fe89846e9881b02fa1df4108f4ae39946acbcdf6ed8bf"
+          "bytes": 13686,
+          "sha256": "a2a35bd3b05e228c99c5ef7cc002c84708dcaff893cd01d51b011dadbf2c02c4"
         }
       },
       "failures": []
@@ -230,27 +226,27 @@
       "local": {
         "path": "repo:docs/data/foundation_model_plan.json",
         "exists": true,
-        "bytes": 8883,
-        "sha256": "0465df50071ab4c8e46acafdd2fa1653f23dacd33a1e1e1d65a571c82348ddbc"
+        "bytes": 8880,
+        "sha256": "a11476cc697e7bc2c29b3db569d059c8f536025fdd0b47e49f060a88664fd450"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/foundation_model_plan.json",
           "exists": true,
-          "bytes": 8883,
-          "sha256": "0465df50071ab4c8e46acafdd2fa1653f23dacd33a1e1e1d65a571c82348ddbc"
+          "bytes": 8880,
+          "sha256": "a11476cc697e7bc2c29b3db569d059c8f536025fdd0b47e49f060a88664fd450"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/foundation_model_plan.json",
           "exists": true,
-          "bytes": 8883,
-          "sha256": "0465df50071ab4c8e46acafdd2fa1653f23dacd33a1e1e1d65a571c82348ddbc"
+          "bytes": 8880,
+          "sha256": "a11476cc697e7bc2c29b3db569d059c8f536025fdd0b47e49f060a88664fd450"
         },
         "hf_model": {
           "path": "hf_model:metrics/foundation_model_plan.json",
           "exists": true,
-          "bytes": 8883,
-          "sha256": "0465df50071ab4c8e46acafdd2fa1653f23dacd33a1e1e1d65a571c82348ddbc"
+          "bytes": 8880,
+          "sha256": "a11476cc697e7bc2c29b3db569d059c8f536025fdd0b47e49f060a88664fd450"
         }
       },
       "failures": []
@@ -261,27 +257,27 @@
       "local": {
         "path": "repo:docs/data/live_publication_status.json",
         "exists": true,
-        "bytes": 68744,
-        "sha256": "9778f6a9dfb926c42c02d7b9d9740d38b970ca7b6ac7bebe735fb33c27ee07de"
+        "bytes": 68749,
+        "sha256": "1095bd7aeee4172afddbe7b9e89c5371605ef22fab294b3d77ec98d5ad7ef0cd"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/live_publication_status.json",
           "exists": true,
-          "bytes": 68744,
-          "sha256": "9778f6a9dfb926c42c02d7b9d9740d38b970ca7b6ac7bebe735fb33c27ee07de"
+          "bytes": 68749,
+          "sha256": "1095bd7aeee4172afddbe7b9e89c5371605ef22fab294b3d77ec98d5ad7ef0cd"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/live_publication_status.json",
           "exists": true,
-          "bytes": 68744,
-          "sha256": "9778f6a9dfb926c42c02d7b9d9740d38b970ca7b6ac7bebe735fb33c27ee07de"
+          "bytes": 68749,
+          "sha256": "1095bd7aeee4172afddbe7b9e89c5371605ef22fab294b3d77ec98d5ad7ef0cd"
         },
         "hf_model": {
           "path": "hf_model:metrics/live_publication_status.json",
           "exists": true,
-          "bytes": 68744,
-          "sha256": "9778f6a9dfb926c42c02d7b9d9740d38b970ca7b6ac7bebe735fb33c27ee07de"
+          "bytes": 68749,
+          "sha256": "1095bd7aeee4172afddbe7b9e89c5371605ef22fab294b3d77ec98d5ad7ef0cd"
         }
       },
       "failures": []
@@ -292,27 +288,27 @@
       "local": {
         "path": "repo:docs/data/modality_atlas.json",
         "exists": true,
-        "bytes": 3819,
-        "sha256": "b1285148588ef402b70c830b62d66fbbcfa86bc5ffc2ad99e165a1429bb49084"
+        "bytes": 3797,
+        "sha256": "4f12a5471d4fe53b1f5c1ef6f9535cc14974690fa6ffbfbeddc8e1b90f86b281"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/modality_atlas.json",
           "exists": true,
-          "bytes": 3819,
-          "sha256": "b1285148588ef402b70c830b62d66fbbcfa86bc5ffc2ad99e165a1429bb49084"
+          "bytes": 3797,
+          "sha256": "4f12a5471d4fe53b1f5c1ef6f9535cc14974690fa6ffbfbeddc8e1b90f86b281"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/modality_atlas.json",
           "exists": true,
-          "bytes": 3819,
-          "sha256": "b1285148588ef402b70c830b62d66fbbcfa86bc5ffc2ad99e165a1429bb49084"
+          "bytes": 3797,
+          "sha256": "4f12a5471d4fe53b1f5c1ef6f9535cc14974690fa6ffbfbeddc8e1b90f86b281"
         },
         "hf_model": {
           "path": "hf_model:metrics/modality_atlas.json",
           "exists": true,
-          "bytes": 3819,
-          "sha256": "b1285148588ef402b70c830b62d66fbbcfa86bc5ffc2ad99e165a1429bb49084"
+          "bytes": 3797,
+          "sha256": "4f12a5471d4fe53b1f5c1ef6f9535cc14974690fa6ffbfbeddc8e1b90f86b281"
         }
       },
       "failures": []
@@ -323,27 +319,27 @@
       "local": {
         "path": "repo:docs/data/project_brief.json",
         "exists": true,
-        "bytes": 2563,
-        "sha256": "c8df06022de6a0d51be993175ac63764a607f48c944f1d9451c5bbeebf6cef54"
+        "bytes": 2585,
+        "sha256": "f0e2e7853b86cdcd2f06269d75f19a96432a782efdf99c08aca7d689fd2cd0a1"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/project_brief.json",
           "exists": true,
-          "bytes": 2563,
-          "sha256": "c8df06022de6a0d51be993175ac63764a607f48c944f1d9451c5bbeebf6cef54"
+          "bytes": 2585,
+          "sha256": "f0e2e7853b86cdcd2f06269d75f19a96432a782efdf99c08aca7d689fd2cd0a1"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/project_brief.json",
           "exists": true,
-          "bytes": 2563,
-          "sha256": "c8df06022de6a0d51be993175ac63764a607f48c944f1d9451c5bbeebf6cef54"
+          "bytes": 2585,
+          "sha256": "f0e2e7853b86cdcd2f06269d75f19a96432a782efdf99c08aca7d689fd2cd0a1"
         },
         "hf_model": {
           "path": "hf_model:metrics/project_brief.json",
           "exists": true,
-          "bytes": 2563,
-          "sha256": "c8df06022de6a0d51be993175ac63764a607f48c944f1d9451c5bbeebf6cef54"
+          "bytes": 2585,
+          "sha256": "f0e2e7853b86cdcd2f06269d75f19a96432a782efdf99c08aca7d689fd2cd0a1"
         }
       },
       "failures": []
@@ -385,27 +381,27 @@
       "local": {
         "path": "repo:docs/data/project_packet.json",
         "exists": true,
-        "bytes": 7659,
-        "sha256": "7d9f5ecc8adf9b3913e99a8aec28372529e63c105739fa53b3f8a7a9470470c2"
+        "bytes": 7627,
+        "sha256": "9b36df95df2a391b93282bbf465dca1594fe3edac96bad128aa331b0b835921a"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/project_packet.json",
           "exists": true,
-          "bytes": 7659,
-          "sha256": "7d9f5ecc8adf9b3913e99a8aec28372529e63c105739fa53b3f8a7a9470470c2"
+          "bytes": 7627,
+          "sha256": "9b36df95df2a391b93282bbf465dca1594fe3edac96bad128aa331b0b835921a"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/project_packet.json",
           "exists": true,
-          "bytes": 7659,
-          "sha256": "7d9f5ecc8adf9b3913e99a8aec28372529e63c105739fa53b3f8a7a9470470c2"
+          "bytes": 7627,
+          "sha256": "9b36df95df2a391b93282bbf465dca1594fe3edac96bad128aa331b0b835921a"
         },
         "hf_model": {
           "path": "hf_model:metrics/project_packet.json",
           "exists": true,
-          "bytes": 7659,
-          "sha256": "7d9f5ecc8adf9b3913e99a8aec28372529e63c105739fa53b3f8a7a9470470c2"
+          "bytes": 7627,
+          "sha256": "9b36df95df2a391b93282bbf465dca1594fe3edac96bad128aa331b0b835921a"
         }
       },
       "failures": []
@@ -416,27 +412,27 @@
       "local": {
         "path": "repo:docs/data/project_status.json",
         "exists": true,
-        "bytes": 9006,
-        "sha256": "38d289ce14e7b4f0998ce156718662f2298ee3aa4e77782a40316ad33aefaefb"
+        "bytes": 8930,
+        "sha256": "b4f0fd7644009ff4baeb98c05beb6b8bca86b166fdb6a7dc4f263a61a7ca13dd"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/project_status.json",
           "exists": true,
-          "bytes": 9006,
-          "sha256": "38d289ce14e7b4f0998ce156718662f2298ee3aa4e77782a40316ad33aefaefb"
+          "bytes": 8930,
+          "sha256": "b4f0fd7644009ff4baeb98c05beb6b8bca86b166fdb6a7dc4f263a61a7ca13dd"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/project_status.json",
           "exists": true,
-          "bytes": 9006,
-          "sha256": "38d289ce14e7b4f0998ce156718662f2298ee3aa4e77782a40316ad33aefaefb"
+          "bytes": 8930,
+          "sha256": "b4f0fd7644009ff4baeb98c05beb6b8bca86b166fdb6a7dc4f263a61a7ca13dd"
         },
         "hf_model": {
           "path": "hf_model:metrics/project_status.json",
           "exists": true,
-          "bytes": 9006,
-          "sha256": "38d289ce14e7b4f0998ce156718662f2298ee3aa4e77782a40316ad33aefaefb"
+          "bytes": 8930,
+          "sha256": "b4f0fd7644009ff4baeb98c05beb6b8bca86b166fdb6a7dc4f263a61a7ca13dd"
         }
       },
       "failures": []
@@ -448,26 +444,26 @@
         "path": "repo:docs/data/publication_audit.json",
         "exists": true,
         "bytes": 7289,
-        "sha256": "5cd9531d57ee355513eb498a73b71b8d0db8dd17f8589dfbf78bc9546d5a3874"
+        "sha256": "a019468bf2be313087ffa4b11088ccc633ac0482be38b2bf13e23540364be466"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/publication_audit.json",
           "exists": true,
           "bytes": 7289,
-          "sha256": "5cd9531d57ee355513eb498a73b71b8d0db8dd17f8589dfbf78bc9546d5a3874"
+          "sha256": "a019468bf2be313087ffa4b11088ccc633ac0482be38b2bf13e23540364be466"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/publication_audit.json",
           "exists": true,
           "bytes": 7289,
-          "sha256": "5cd9531d57ee355513eb498a73b71b8d0db8dd17f8589dfbf78bc9546d5a3874"
+          "sha256": "a019468bf2be313087ffa4b11088ccc633ac0482be38b2bf13e23540364be466"
         },
         "hf_model": {
           "path": "hf_model:metrics/publication_audit.json",
           "exists": true,
           "bytes": 7289,
-          "sha256": "5cd9531d57ee355513eb498a73b71b8d0db8dd17f8589dfbf78bc9546d5a3874"
+          "sha256": "a019468bf2be313087ffa4b11088ccc633ac0482be38b2bf13e23540364be466"
         }
       },
       "failures": []
@@ -479,26 +475,26 @@
         "path": "repo:docs/data/public_surface_qa.json",
         "exists": true,
         "bytes": 5631,
-        "sha256": "74286fc8c887055f1d4b41d0e4b9dae552f733c01240eeeb1f09bb5e47205446"
+        "sha256": "0f61bee108f0dc72f48c4edd0ca194fc5d108bb0c5cb7bcc86c6eaa56f5fae64"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/public_surface_qa.json",
           "exists": true,
           "bytes": 5631,
-          "sha256": "74286fc8c887055f1d4b41d0e4b9dae552f733c01240eeeb1f09bb5e47205446"
+          "sha256": "0f61bee108f0dc72f48c4edd0ca194fc5d108bb0c5cb7bcc86c6eaa56f5fae64"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/public_surface_qa.json",
           "exists": true,
           "bytes": 5631,
-          "sha256": "74286fc8c887055f1d4b41d0e4b9dae552f733c01240eeeb1f09bb5e47205446"
+          "sha256": "0f61bee108f0dc72f48c4edd0ca194fc5d108bb0c5cb7bcc86c6eaa56f5fae64"
         },
         "hf_model": {
           "path": "hf_model:metrics/public_surface_qa.json",
           "exists": true,
           "bytes": 5631,
-          "sha256": "74286fc8c887055f1d4b41d0e4b9dae552f733c01240eeeb1f09bb5e47205446"
+          "sha256": "0f61bee108f0dc72f48c4edd0ca194fc5d108bb0c5cb7bcc86c6eaa56f5fae64"
         }
       },
       "failures": []
@@ -509,27 +505,27 @@
       "local": {
         "path": "repo:docs/data/quality_gates.json",
         "exists": true,
-        "bytes": 8147,
-        "sha256": "1d4d45937e7d90ee5d54125e4d5d7b5d5eee4878251b52152e59a34efc557c04"
+        "bytes": 8118,
+        "sha256": "595b2e050028b612b57ecfb5086c66bdf5040be90250430537cbd9e1b9c09e76"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/quality_gates.json",
           "exists": true,
-          "bytes": 8147,
-          "sha256": "1d4d45937e7d90ee5d54125e4d5d7b5d5eee4878251b52152e59a34efc557c04"
+          "bytes": 8118,
+          "sha256": "595b2e050028b612b57ecfb5086c66bdf5040be90250430537cbd9e1b9c09e76"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/quality_gates.json",
           "exists": true,
-          "bytes": 8147,
-          "sha256": "1d4d45937e7d90ee5d54125e4d5d7b5d5eee4878251b52152e59a34efc557c04"
+          "bytes": 8118,
+          "sha256": "595b2e050028b612b57ecfb5086c66bdf5040be90250430537cbd9e1b9c09e76"
         },
         "hf_model": {
           "path": "hf_model:metrics/quality_gates.json",
           "exists": true,
-          "bytes": 8147,
-          "sha256": "1d4d45937e7d90ee5d54125e4d5d7b5d5eee4878251b52152e59a34efc557c04"
+          "bytes": 8118,
+          "sha256": "595b2e050028b612b57ecfb5086c66bdf5040be90250430537cbd9e1b9c09e76"
         }
       },
       "failures": []
@@ -571,27 +567,27 @@
       "local": {
         "path": "repo:docs/data/reproducibility_matrix.json",
         "exists": true,
-        "bytes": 5197,
-        "sha256": "c3d353084ee641ff3d1369dd911577ff7a1b5d7b2faf841faa12caa69dfa5ed0"
+        "bytes": 5229,
+        "sha256": "928eb95b95d4028a7f21135584c37ff141833e26178cc918a874a6448f8c5ded"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5197,
-          "sha256": "c3d353084ee641ff3d1369dd911577ff7a1b5d7b2faf841faa12caa69dfa5ed0"
+          "bytes": 5229,
+          "sha256": "928eb95b95d4028a7f21135584c37ff141833e26178cc918a874a6448f8c5ded"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5197,
-          "sha256": "c3d353084ee641ff3d1369dd911577ff7a1b5d7b2faf841faa12caa69dfa5ed0"
+          "bytes": 5229,
+          "sha256": "928eb95b95d4028a7f21135584c37ff141833e26178cc918a874a6448f8c5ded"
         },
         "hf_model": {
           "path": "hf_model:metrics/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5197,
-          "sha256": "c3d353084ee641ff3d1369dd911577ff7a1b5d7b2faf841faa12caa69dfa5ed0"
+          "bytes": 5229,
+          "sha256": "928eb95b95d4028a7f21135584c37ff141833e26178cc918a874a6448f8c5ded"
         }
       },
       "failures": []
@@ -602,27 +598,27 @@
       "local": {
         "path": "repo:docs/data/research_roadmap.json",
         "exists": true,
-        "bytes": 5752,
-        "sha256": "0510e919ad9ffa3aebe052273e90133a57572c3b21b991db1ff28de83061f6ca"
+        "bytes": 5726,
+        "sha256": "903ec0d6104c092fa0356c7741134fee4a1c78f3b0517001c49ce11f8077d6e2"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/research_roadmap.json",
           "exists": true,
-          "bytes": 5752,
-          "sha256": "0510e919ad9ffa3aebe052273e90133a57572c3b21b991db1ff28de83061f6ca"
+          "bytes": 5726,
+          "sha256": "903ec0d6104c092fa0356c7741134fee4a1c78f3b0517001c49ce11f8077d6e2"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/research_roadmap.json",
           "exists": true,
-          "bytes": 5752,
-          "sha256": "0510e919ad9ffa3aebe052273e90133a57572c3b21b991db1ff28de83061f6ca"
+          "bytes": 5726,
+          "sha256": "903ec0d6104c092fa0356c7741134fee4a1c78f3b0517001c49ce11f8077d6e2"
         },
         "hf_model": {
           "path": "hf_model:metrics/research_roadmap.json",
           "exists": true,
-          "bytes": 5752,
-          "sha256": "0510e919ad9ffa3aebe052273e90133a57572c3b21b991db1ff28de83061f6ca"
+          "bytes": 5726,
+          "sha256": "903ec0d6104c092fa0356c7741134fee4a1c78f3b0517001c49ce11f8077d6e2"
         }
       },
       "failures": []
@@ -633,27 +629,27 @@
       "local": {
         "path": "repo:docs/data/research_roadmap_interactive.json",
         "exists": true,
-        "bytes": 131526,
-        "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0"
+        "bytes": 131473,
+        "sha256": "c0e2f4872c7192995307967d27bf9c329ae88c685688e3f128f68e3fdc48e780"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/research_roadmap_interactive.json",
           "exists": true,
-          "bytes": 131526,
-          "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0"
+          "bytes": 131473,
+          "sha256": "c0e2f4872c7192995307967d27bf9c329ae88c685688e3f128f68e3fdc48e780"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/research_roadmap_interactive.json",
           "exists": true,
-          "bytes": 131526,
-          "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0"
+          "bytes": 131473,
+          "sha256": "c0e2f4872c7192995307967d27bf9c329ae88c685688e3f128f68e3fdc48e780"
         },
         "hf_model": {
           "path": "hf_model:metrics/research_roadmap_interactive.json",
           "exists": true,
-          "bytes": 131526,
-          "sha256": "12b64fa9108066664839c8bc0b02e3c5ddb6d6b10311fb96f8bfc265f25d3cc0"
+          "bytes": 131473,
+          "sha256": "c0e2f4872c7192995307967d27bf9c329ae88c685688e3f128f68e3fdc48e780"
         }
       },
       "failures": []
@@ -664,27 +660,27 @@
       "local": {
         "path": "repo:docs/data/research_takeaways.json",
         "exists": true,
-        "bytes": 6814,
-        "sha256": "538ec80950dbf5d77cc6ab2bd5a96a3af191b56a257b895e48b6cf28deb0d044"
+        "bytes": 6821,
+        "sha256": "509d8690a4854097ea45af635dafa0231efa11838f1bb1f9cb338e99009a0206"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/research_takeaways.json",
           "exists": true,
-          "bytes": 6814,
-          "sha256": "538ec80950dbf5d77cc6ab2bd5a96a3af191b56a257b895e48b6cf28deb0d044"
+          "bytes": 6821,
+          "sha256": "509d8690a4854097ea45af635dafa0231efa11838f1bb1f9cb338e99009a0206"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/research_takeaways.json",
           "exists": true,
-          "bytes": 6814,
-          "sha256": "538ec80950dbf5d77cc6ab2bd5a96a3af191b56a257b895e48b6cf28deb0d044"
+          "bytes": 6821,
+          "sha256": "509d8690a4854097ea45af635dafa0231efa11838f1bb1f9cb338e99009a0206"
         },
         "hf_model": {
           "path": "hf_model:metrics/research_takeaways.json",
           "exists": true,
-          "bytes": 6814,
-          "sha256": "538ec80950dbf5d77cc6ab2bd5a96a3af191b56a257b895e48b6cf28deb0d044"
+          "bytes": 6821,
+          "sha256": "509d8690a4854097ea45af635dafa0231efa11838f1bb1f9cb338e99009a0206"
         }
       },
       "failures": []
@@ -695,27 +691,27 @@
       "local": {
         "path": "repo:docs/data/research_direction_extensions.json",
         "exists": true,
-        "bytes": 11907,
-        "sha256": "aa65022a8a41af4f0bca5197add932c4f3f51b13b8324323280f64457473ff03"
+        "bytes": 11903,
+        "sha256": "010e254b61de6cc199e8e710fb10d8304c8156835f5cbafb79202996a74c0c77"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/research_direction_extensions.json",
           "exists": true,
-          "bytes": 11907,
-          "sha256": "aa65022a8a41af4f0bca5197add932c4f3f51b13b8324323280f64457473ff03"
+          "bytes": 11903,
+          "sha256": "010e254b61de6cc199e8e710fb10d8304c8156835f5cbafb79202996a74c0c77"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/research_direction_extensions.json",
           "exists": true,
-          "bytes": 11907,
-          "sha256": "aa65022a8a41af4f0bca5197add932c4f3f51b13b8324323280f64457473ff03"
+          "bytes": 11903,
+          "sha256": "010e254b61de6cc199e8e710fb10d8304c8156835f5cbafb79202996a74c0c77"
         },
         "hf_model": {
           "path": "hf_model:metrics/research_direction_extensions.json",
           "exists": true,
-          "bytes": 11907,
-          "sha256": "aa65022a8a41af4f0bca5197add932c4f3f51b13b8324323280f64457473ff03"
+          "bytes": 11903,
+          "sha256": "010e254b61de6cc199e8e710fb10d8304c8156835f5cbafb79202996a74c0c77"
         }
       },
       "failures": []
@@ -726,27 +722,27 @@
       "local": {
         "path": "repo:docs/data/research_directions.json",
         "exists": true,
-        "bytes": 14429,
-        "sha256": "292ddd47b5f50cc5a6a0d4099466b134774a1ca7610afe50a2a5781c6c181bf9"
+        "bytes": 14414,
+        "sha256": "7215681c55a6739da7f16d833f62fcb8f6d58069840963182e98427eaf0cf654"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/research_directions.json",
           "exists": true,
-          "bytes": 14429,
-          "sha256": "292ddd47b5f50cc5a6a0d4099466b134774a1ca7610afe50a2a5781c6c181bf9"
+          "bytes": 14414,
+          "sha256": "7215681c55a6739da7f16d833f62fcb8f6d58069840963182e98427eaf0cf654"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/research_directions.json",
           "exists": true,
-          "bytes": 14429,
-          "sha256": "292ddd47b5f50cc5a6a0d4099466b134774a1ca7610afe50a2a5781c6c181bf9"
+          "bytes": 14414,
+          "sha256": "7215681c55a6739da7f16d833f62fcb8f6d58069840963182e98427eaf0cf654"
         },
         "hf_model": {
           "path": "hf_model:metrics/research_directions.json",
           "exists": true,
-          "bytes": 14429,
-          "sha256": "292ddd47b5f50cc5a6a0d4099466b134774a1ca7610afe50a2a5781c6c181bf9"
+          "bytes": 14414,
+          "sha256": "7215681c55a6739da7f16d833f62fcb8f6d58069840963182e98427eaf0cf654"
         }
       },
       "failures": []
@@ -815,56 +811,34 @@
     },
     {
       "name": "data/source_alignment_audit.json",
-      "status": "fail",
+      "status": "pass",
       "local": {
         "path": "repo:docs/data/source_alignment_audit.json",
         "exists": true,
-        "bytes": 4874,
-        "sha256": "d7624ff175e25efdbee2004dfc328269dda7d55c19469f2b18c2b5bdafee0f36"
+        "bytes": 4432,
+        "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
         },
         "hf_model": {
           "path": "hf_model:metrics/source_alignment_audit.json",
           "exists": true,
           "bytes": 4432,
-          "sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
+          "sha256": "e05e785b906d7f0ccedc6ea3bb563eb8dd923f237094f4556d2ea351178f4b5b"
         }
       },
-      "failures": [
-        {
-          "surface": "hf_space",
-          "kind": "hash_mismatch",
-          "path": "hf_space:data/source_alignment_audit.json",
-          "expected_sha256": "d7624ff175e25efdbee2004dfc328269dda7d55c19469f2b18c2b5bdafee0f36",
-          "actual_sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
-        },
-        {
-          "surface": "hf_artifacts",
-          "kind": "hash_mismatch",
-          "path": "hf_artifacts:docs/data/source_alignment_audit.json",
-          "expected_sha256": "d7624ff175e25efdbee2004dfc328269dda7d55c19469f2b18c2b5bdafee0f36",
-          "actual_sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
-        },
-        {
-          "surface": "hf_model",
-          "kind": "hash_mismatch",
-          "path": "hf_model:metrics/source_alignment_audit.json",
-          "expected_sha256": "d7624ff175e25efdbee2004dfc328269dda7d55c19469f2b18c2b5bdafee0f36",
-          "actual_sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
-        }
-      ]
+      "failures": []
     },
     {
       "name": "data/summary_metrics.json",
@@ -872,27 +846,27 @@
       "local": {
         "path": "repo:docs/data/summary_metrics.json",
         "exists": true,
-        "bytes": 25210,
-        "sha256": "a88302490b9099e7d148a8a6dd0ba6393d7ef26a0ba2c6a0edf552da94b948a8"
+        "bytes": 25260,
+        "sha256": "e53e0a6151df2dd0bf02467421abf5a8e7655b88a21b6ce984786adca5d4cc43"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/summary_metrics.json",
           "exists": true,
-          "bytes": 25210,
-          "sha256": "a88302490b9099e7d148a8a6dd0ba6393d7ef26a0ba2c6a0edf552da94b948a8"
+          "bytes": 25260,
+          "sha256": "e53e0a6151df2dd0bf02467421abf5a8e7655b88a21b6ce984786adca5d4cc43"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/summary_metrics.json",
           "exists": true,
-          "bytes": 25210,
-          "sha256": "a88302490b9099e7d148a8a6dd0ba6393d7ef26a0ba2c6a0edf552da94b948a8"
+          "bytes": 25260,
+          "sha256": "e53e0a6151df2dd0bf02467421abf5a8e7655b88a21b6ce984786adca5d4cc43"
         },
         "hf_model": {
           "path": "hf_model:metrics/summary_metrics.json",
           "exists": true,
-          "bytes": 25210,
-          "sha256": "a88302490b9099e7d148a8a6dd0ba6393d7ef26a0ba2c6a0edf552da94b948a8"
+          "bytes": 25260,
+          "sha256": "e53e0a6151df2dd0bf02467421abf5a8e7655b88a21b6ce984786adca5d4cc43"
         }
       },
       "failures": []
@@ -934,27 +908,27 @@
       "local": {
         "path": "repo:docs/data/task_walkthroughs.json",
         "exists": true,
-        "bytes": 26976,
-        "sha256": "4a43c4df17f26cd02c988afab34165a9b106acf440cf5a09d9f042b6e1fada1b"
+        "bytes": 26963,
+        "sha256": "b93df37e7da1fcd5d3df7c2643b4659e2fef986dba52cb37a164874919ccc53a"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/task_walkthroughs.json",
           "exists": true,
-          "bytes": 26976,
-          "sha256": "4a43c4df17f26cd02c988afab34165a9b106acf440cf5a09d9f042b6e1fada1b"
+          "bytes": 26963,
+          "sha256": "b93df37e7da1fcd5d3df7c2643b4659e2fef986dba52cb37a164874919ccc53a"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/task_walkthroughs.json",
           "exists": true,
-          "bytes": 26976,
-          "sha256": "4a43c4df17f26cd02c988afab34165a9b106acf440cf5a09d9f042b6e1fada1b"
+          "bytes": 26963,
+          "sha256": "b93df37e7da1fcd5d3df7c2643b4659e2fef986dba52cb37a164874919ccc53a"
         },
         "hf_model": {
           "path": "hf_model:metrics/task_walkthroughs.json",
           "exists": true,
-          "bytes": 26976,
-          "sha256": "4a43c4df17f26cd02c988afab34165a9b106acf440cf5a09d9f042b6e1fada1b"
+          "bytes": 26963,
+          "sha256": "b93df37e7da1fcd5d3df7c2643b4659e2fef986dba52cb37a164874919ccc53a"
         }
       },
       "failures": []
@@ -966,26 +940,26 @@
         "path": "repo:docs/data/website_integrity.json",
         "exists": true,
         "bytes": 14718,
-        "sha256": "ddd842cd2692bbe1a029777ff91514dd611413305b97c0f6fcce31cd668c61b7"
+        "sha256": "96372b1d3297dd81a393010264d4d92204bd140838aa7a81f227fea7d88b5fb9"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/website_integrity.json",
           "exists": true,
           "bytes": 14718,
-          "sha256": "ddd842cd2692bbe1a029777ff91514dd611413305b97c0f6fcce31cd668c61b7"
+          "sha256": "96372b1d3297dd81a393010264d4d92204bd140838aa7a81f227fea7d88b5fb9"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/website_integrity.json",
           "exists": true,
           "bytes": 14718,
-          "sha256": "ddd842cd2692bbe1a029777ff91514dd611413305b97c0f6fcce31cd668c61b7"
+          "sha256": "96372b1d3297dd81a393010264d4d92204bd140838aa7a81f227fea7d88b5fb9"
         },
         "hf_model": {
           "path": "hf_model:metrics/website_integrity.json",
           "exists": true,
           "bytes": 14718,
-          "sha256": "ddd842cd2692bbe1a029777ff91514dd611413305b97c0f6fcce31cd668c61b7"
+          "sha256": "96372b1d3297dd81a393010264d4d92204bd140838aa7a81f227fea7d88b5fb9"
         }
       },
       "failures": []
@@ -996,27 +970,27 @@
       "local": {
         "path": "repo:docs/data/xperience10m_dataset_card_alignment.json",
         "exists": true,
-        "bytes": 7573,
-        "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7"
+        "bytes": 7559,
+        "sha256": "24ea3fc9e1ae8f698ca192ae1b8511dfe1129b807af237f41382e0644e5b58af"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/xperience10m_dataset_card_alignment.json",
           "exists": true,
-          "bytes": 7573,
-          "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7"
+          "bytes": 7559,
+          "sha256": "24ea3fc9e1ae8f698ca192ae1b8511dfe1129b807af237f41382e0644e5b58af"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/xperience10m_dataset_card_alignment.json",
           "exists": true,
-          "bytes": 7573,
-          "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7"
+          "bytes": 7559,
+          "sha256": "24ea3fc9e1ae8f698ca192ae1b8511dfe1129b807af237f41382e0644e5b58af"
         },
         "hf_model": {
           "path": "hf_model:metrics/xperience10m_dataset_card_alignment.json",
           "exists": true,
-          "bytes": 7573,
-          "sha256": "a06402b0a9c23c8bacf0e79acd6bd07ba87e555f3c992d573eac17a3569eb2d7"
+          "bytes": 7559,
+          "sha256": "24ea3fc9e1ae8f698ca192ae1b8511dfe1129b807af237f41382e0644e5b58af"
         }
       },
       "failures": []
@@ -1693,21 +1667,21 @@
       "local": {
         "path": "repo:scripts/audio_ablation_and_raw_upgrade.py",
         "exists": true,
-        "bytes": 43192,
-        "sha256": "73db45a3ddb50b60f61ca5e55bdf2008e48c268681bfcdc95d35bb56e48f89e3"
+        "bytes": 43144,
+        "sha256": "f7e3a38ec906dac7ca55b13c49720bd41ed89a1fd994c7d54730a4de5dfd1b59"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/audio_ablation_and_raw_upgrade.py",
           "exists": true,
-          "bytes": 43192,
-          "sha256": "73db45a3ddb50b60f61ca5e55bdf2008e48c268681bfcdc95d35bb56e48f89e3"
+          "bytes": 43144,
+          "sha256": "f7e3a38ec906dac7ca55b13c49720bd41ed89a1fd994c7d54730a4de5dfd1b59"
         },
         "hf_model": {
           "path": "hf_model:scripts/audio_ablation_and_raw_upgrade.py",
           "exists": true,
-          "bytes": 43192,
-          "sha256": "73db45a3ddb50b60f61ca5e55bdf2008e48c268681bfcdc95d35bb56e48f89e3"
+          "bytes": 43144,
+          "sha256": "f7e3a38ec906dac7ca55b13c49720bd41ed89a1fd994c7d54730a4de5dfd1b59"
         }
       },
       "failures": []
@@ -1718,21 +1692,21 @@
       "local": {
         "path": "repo:scripts/build_artifact_index.py",
         "exists": true,
-        "bytes": 26652,
-        "sha256": "faf1e0b56eb9ba4bc94e04bb1aeb9a392fb4410e8053564dd43bce3d8aacfa4d"
+        "bytes": 26566,
+        "sha256": "982fcc36e6dfda3382662d1f1e1f539f27f45aaddd6503aad77a14145bcc1bf6"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_artifact_index.py",
           "exists": true,
-          "bytes": 26652,
-          "sha256": "faf1e0b56eb9ba4bc94e04bb1aeb9a392fb4410e8053564dd43bce3d8aacfa4d"
+          "bytes": 26566,
+          "sha256": "982fcc36e6dfda3382662d1f1e1f539f27f45aaddd6503aad77a14145bcc1bf6"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_artifact_index.py",
           "exists": true,
-          "bytes": 26652,
-          "sha256": "faf1e0b56eb9ba4bc94e04bb1aeb9a392fb4410e8053564dd43bce3d8aacfa4d"
+          "bytes": 26566,
+          "sha256": "982fcc36e6dfda3382662d1f1e1f539f27f45aaddd6503aad77a14145bcc1bf6"
         }
       },
       "failures": []
@@ -1768,21 +1742,21 @@
       "local": {
         "path": "repo:scripts/build_evaluation_protocol.py",
         "exists": true,
-        "bytes": 16163,
-        "sha256": "b352798b50212dac792855313bdc0ad1e5e2ef835de1ff25b03d52afd1d33f03"
+        "bytes": 16139,
+        "sha256": "e2aaac38f451402971e03c004d32d5c5f8a05d67dc1820d6bf9b12d1da1a292f"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_evaluation_protocol.py",
           "exists": true,
-          "bytes": 16163,
-          "sha256": "b352798b50212dac792855313bdc0ad1e5e2ef835de1ff25b03d52afd1d33f03"
+          "bytes": 16139,
+          "sha256": "e2aaac38f451402971e03c004d32d5c5f8a05d67dc1820d6bf9b12d1da1a292f"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_evaluation_protocol.py",
           "exists": true,
-          "bytes": 16163,
-          "sha256": "b352798b50212dac792855313bdc0ad1e5e2ef835de1ff25b03d52afd1d33f03"
+          "bytes": 16139,
+          "sha256": "e2aaac38f451402971e03c004d32d5c5f8a05d67dc1820d6bf9b12d1da1a292f"
         }
       },
       "failures": []
@@ -1818,21 +1792,21 @@
       "local": {
         "path": "repo:scripts/build_quality_gates.py",
         "exists": true,
-        "bytes": 11380,
-        "sha256": "e8c99db9b2698f824b580882b238a881cffeafffc0b5f768d5fad0f9ac24ae18"
+        "bytes": 11351,
+        "sha256": "1a185999e568243fb480ba21d1fc181037a97136dec4e562b3b78fb1dd7d9ead"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_quality_gates.py",
           "exists": true,
-          "bytes": 11380,
-          "sha256": "e8c99db9b2698f824b580882b238a881cffeafffc0b5f768d5fad0f9ac24ae18"
+          "bytes": 11351,
+          "sha256": "1a185999e568243fb480ba21d1fc181037a97136dec4e562b3b78fb1dd7d9ead"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_quality_gates.py",
           "exists": true,
-          "bytes": 11380,
-          "sha256": "e8c99db9b2698f824b580882b238a881cffeafffc0b5f768d5fad0f9ac24ae18"
+          "bytes": 11351,
+          "sha256": "1a185999e568243fb480ba21d1fc181037a97136dec4e562b3b78fb1dd7d9ead"
         }
       },
       "failures": []
@@ -1918,21 +1892,21 @@
       "local": {
         "path": "repo:scripts/build_single_episode_explorer.py",
         "exists": true,
-        "bytes": 30204,
-        "sha256": "e39890ddc960605f2f667b53672153cac572b165b8a2a7d8518d6e0777788762"
+        "bytes": 30200,
+        "sha256": "57924e42243c76fb37dc36acb1c41932edd0467da7b2794ea6bd08603884b61a"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_single_episode_explorer.py",
           "exists": true,
-          "bytes": 30204,
-          "sha256": "e39890ddc960605f2f667b53672153cac572b165b8a2a7d8518d6e0777788762"
+          "bytes": 30200,
+          "sha256": "57924e42243c76fb37dc36acb1c41932edd0467da7b2794ea6bd08603884b61a"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_single_episode_explorer.py",
           "exists": true,
-          "bytes": 30204,
-          "sha256": "e39890ddc960605f2f667b53672153cac572b165b8a2a7d8518d6e0777788762"
+          "bytes": 30200,
+          "sha256": "57924e42243c76fb37dc36acb1c41932edd0467da7b2794ea6bd08603884b61a"
         }
       },
       "failures": []
@@ -1943,21 +1917,21 @@
       "local": {
         "path": "repo:scripts/build_research_takeaways.py",
         "exists": true,
-        "bytes": 13304,
-        "sha256": "135f555b17aae5dc5965ec88c3db6e49622a480467909ffa10d53278d4ea054b"
+        "bytes": 13285,
+        "sha256": "19baa1dbe3e4454e040bb1b2fc711e1dd5ce5280ee19bfc01b01cca854006d38"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/build_research_takeaways.py",
           "exists": true,
-          "bytes": 13304,
-          "sha256": "135f555b17aae5dc5965ec88c3db6e49622a480467909ffa10d53278d4ea054b"
+          "bytes": 13285,
+          "sha256": "19baa1dbe3e4454e040bb1b2fc711e1dd5ce5280ee19bfc01b01cca854006d38"
         },
         "hf_model": {
           "path": "hf_model:scripts/build_research_takeaways.py",
           "exists": true,
-          "bytes": 13304,
-          "sha256": "135f555b17aae5dc5965ec88c3db6e49622a480467909ffa10d53278d4ea054b"
+          "bytes": 13285,
+          "sha256": "19baa1dbe3e4454e040bb1b2fc711e1dd5ce5280ee19bfc01b01cca854006d38"
         }
       },
       "failures": []
@@ -1993,21 +1967,21 @@
       "local": {
         "path": "repo:scripts/verify_live_publication.py",
         "exists": true,
-        "bytes": 34767,
-        "sha256": "6a1cb0531719888de6bbeb7a2628058f6e67c92df6314c4ac3952af2c17f126b"
+        "bytes": 34779,
+        "sha256": "73c45e1ae1b7509b9f2c54cf42424c9ac0aab8a9e54be58fc488eaa2a696acc7"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/verify_live_publication.py",
           "exists": true,
-          "bytes": 34767,
-          "sha256": "6a1cb0531719888de6bbeb7a2628058f6e67c92df6314c4ac3952af2c17f126b"
+          "bytes": 34779,
+          "sha256": "73c45e1ae1b7509b9f2c54cf42424c9ac0aab8a9e54be58fc488eaa2a696acc7"
         },
         "hf_model": {
           "path": "hf_model:scripts/verify_live_publication.py",
           "exists": true,
-          "bytes": 34767,
-          "sha256": "6a1cb0531719888de6bbeb7a2628058f6e67c92df6314c4ac3952af2c17f126b"
+          "bytes": 34779,
+          "sha256": "73c45e1ae1b7509b9f2c54cf42424c9ac0aab8a9e54be58fc488eaa2a696acc7"
         }
       },
       "failures": []
@@ -2043,21 +2017,21 @@
       "local": {
         "path": "repo:scripts/validate_publication_package.py",
         "exists": true,
-        "bytes": 19247,
-        "sha256": "55a9c51d57766d54f9eb884e06d644fd9f3d6e358b80c06e7f515ab764cc2158"
+        "bytes": 19276,
+        "sha256": "861b599ed7df030942e179cb8e9e68b5395a11e97d01d2775ecd5c9c2a3916d2"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/validate_publication_package.py",
           "exists": true,
-          "bytes": 19247,
-          "sha256": "55a9c51d57766d54f9eb884e06d644fd9f3d6e358b80c06e7f515ab764cc2158"
+          "bytes": 19276,
+          "sha256": "861b599ed7df030942e179cb8e9e68b5395a11e97d01d2775ecd5c9c2a3916d2"
         },
         "hf_model": {
           "path": "hf_model:scripts/validate_publication_package.py",
           "exists": true,
-          "bytes": 19247,
-          "sha256": "55a9c51d57766d54f9eb884e06d644fd9f3d6e358b80c06e7f515ab764cc2158"
+          "bytes": 19276,
+          "sha256": "861b599ed7df030942e179cb8e9e68b5395a11e97d01d2775ecd5c9c2a3916d2"
         }
       },
       "failures": []
@@ -2093,21 +2067,21 @@
       "local": {
         "path": "repo:scripts/validate_source_alignment.py",
         "exists": true,
-        "bytes": 15807,
-        "sha256": "66a9d0eebb4800be5d65f982ef13502b37522cd2a9eae0686c63efc121a52b0a"
+        "bytes": 15831,
+        "sha256": "6f48190ca77dd6ba9dd2f821d5ace7e33bd9b4d9d76adb7237cc51ddb88de239"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/validate_source_alignment.py",
           "exists": true,
-          "bytes": 15807,
-          "sha256": "66a9d0eebb4800be5d65f982ef13502b37522cd2a9eae0686c63efc121a52b0a"
+          "bytes": 15831,
+          "sha256": "6f48190ca77dd6ba9dd2f821d5ace7e33bd9b4d9d76adb7237cc51ddb88de239"
         },
         "hf_model": {
           "path": "hf_model:scripts/validate_source_alignment.py",
           "exists": true,
-          "bytes": 15807,
-          "sha256": "66a9d0eebb4800be5d65f982ef13502b37522cd2a9eae0686c63efc121a52b0a"
+          "bytes": 15831,
+          "sha256": "6f48190ca77dd6ba9dd2f821d5ace7e33bd9b4d9d76adb7237cc51ddb88de239"
         }
       },
       "failures": []
@@ -2243,21 +2217,21 @@
       "local": {
         "path": "repo:docs/index.html",
         "exists": true,
-        "bytes": 170442,
-        "sha256": "a07bbd17dc6f0212c7cbea3b6656b2be9f2af11bd28845ac80b528a9f96b2720"
+        "bytes": 170233,
+        "sha256": "669d8f0e4f278802df86787e4e69efa87e301091bb4c83ac388670c0d456c881"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:index.html",
           "exists": true,
-          "bytes": 170442,
-          "sha256": "a07bbd17dc6f0212c7cbea3b6656b2be9f2af11bd28845ac80b528a9f96b2720"
+          "bytes": 170233,
+          "sha256": "669d8f0e4f278802df86787e4e69efa87e301091bb4c83ac388670c0d456c881"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/index.html",
           "exists": true,
-          "bytes": 170442,
-          "sha256": "a07bbd17dc6f0212c7cbea3b6656b2be9f2af11bd28845ac80b528a9f96b2720"
+          "bytes": 170233,
+          "sha256": "669d8f0e4f278802df86787e4e69efa87e301091bb4c83ac388670c0d456c881"
         }
       },
       "failures": []
@@ -2777,27 +2751,27 @@
       "local": {
         "path": "repo:QUALITY_GATES.md",
         "exists": true,
-        "bytes": 4919,
-        "sha256": "7138e99b116c44f128cd2f749e9d7427e496cb7596d30d54dd44af79be80df81"
+        "bytes": 4901,
+        "sha256": "7426ef148491d7288161a8d13cf31e81d5351ab1c920614a59b06888889c97d3"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:QUALITY_GATES.md",
           "exists": true,
-          "bytes": 4919,
-          "sha256": "7138e99b116c44f128cd2f749e9d7427e496cb7596d30d54dd44af79be80df81"
+          "bytes": 4901,
+          "sha256": "7426ef148491d7288161a8d13cf31e81d5351ab1c920614a59b06888889c97d3"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:QUALITY_GATES.md",
           "exists": true,
-          "bytes": 4919,
-          "sha256": "7138e99b116c44f128cd2f749e9d7427e496cb7596d30d54dd44af79be80df81"
+          "bytes": 4901,
+          "sha256": "7426ef148491d7288161a8d13cf31e81d5351ab1c920614a59b06888889c97d3"
         },
         "hf_model": {
           "path": "hf_model:QUALITY_GATES.md",
           "exists": true,
-          "bytes": 4919,
-          "sha256": "7138e99b116c44f128cd2f749e9d7427e496cb7596d30d54dd44af79be80df81"
+          "bytes": 4901,
+          "sha256": "7426ef148491d7288161a8d13cf31e81d5351ab1c920614a59b06888889c97d3"
         }
       },
       "failures": []
@@ -2870,27 +2844,27 @@
       "local": {
         "path": "repo:FOUNDATION_MODEL_PLAN.md",
         "exists": true,
-        "bytes": 6538,
-        "sha256": "15431b3ed368d660e777655d2e4a97f2870997f2ad38b451a73d631b5e431bcf"
+        "bytes": 6545,
+        "sha256": "f7d35f38db371d52b5bb92e59b951b93946c425fe12bf283f69f869a45a5c60f"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:FOUNDATION_MODEL_PLAN.md",
           "exists": true,
-          "bytes": 6538,
-          "sha256": "15431b3ed368d660e777655d2e4a97f2870997f2ad38b451a73d631b5e431bcf"
+          "bytes": 6545,
+          "sha256": "f7d35f38db371d52b5bb92e59b951b93946c425fe12bf283f69f869a45a5c60f"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:FOUNDATION_MODEL_PLAN.md",
           "exists": true,
-          "bytes": 6538,
-          "sha256": "15431b3ed368d660e777655d2e4a97f2870997f2ad38b451a73d631b5e431bcf"
+          "bytes": 6545,
+          "sha256": "f7d35f38db371d52b5bb92e59b951b93946c425fe12bf283f69f869a45a5c60f"
         },
         "hf_model": {
           "path": "hf_model:FOUNDATION_MODEL_PLAN.md",
           "exists": true,
-          "bytes": 6538,
-          "sha256": "15431b3ed368d660e777655d2e4a97f2870997f2ad38b451a73d631b5e431bcf"
+          "bytes": 6545,
+          "sha256": "f7d35f38db371d52b5bb92e59b951b93946c425fe12bf283f69f869a45a5c60f"
         }
       },
       "failures": []
@@ -2901,27 +2875,27 @@
       "local": {
         "path": "repo:PROJECT_BRIEF.md",
         "exists": true,
-        "bytes": 2494,
-        "sha256": "54c6906103f6d7f45398cd638730c3eb9813d6ca4a807920d6edba4b12f446e6"
+        "bytes": 2516,
+        "sha256": "ace07e60bab3d17d17d2af32b9843582adfbb8d0abe5f7c0489c1e9502eab2f8"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:PROJECT_BRIEF.md",
           "exists": true,
-          "bytes": 2494,
-          "sha256": "54c6906103f6d7f45398cd638730c3eb9813d6ca4a807920d6edba4b12f446e6"
+          "bytes": 2516,
+          "sha256": "ace07e60bab3d17d17d2af32b9843582adfbb8d0abe5f7c0489c1e9502eab2f8"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:PROJECT_BRIEF.md",
           "exists": true,
-          "bytes": 2494,
-          "sha256": "54c6906103f6d7f45398cd638730c3eb9813d6ca4a807920d6edba4b12f446e6"
+          "bytes": 2516,
+          "sha256": "ace07e60bab3d17d17d2af32b9843582adfbb8d0abe5f7c0489c1e9502eab2f8"
         },
         "hf_model": {
           "path": "hf_model:PROJECT_BRIEF.md",
           "exists": true,
-          "bytes": 2494,
-          "sha256": "54c6906103f6d7f45398cd638730c3eb9813d6ca4a807920d6edba4b12f446e6"
+          "bytes": 2516,
+          "sha256": "ace07e60bab3d17d17d2af32b9843582adfbb8d0abe5f7c0489c1e9502eab2f8"
         }
       },
       "failures": []
@@ -2963,27 +2937,27 @@
       "local": {
         "path": "repo:RESEARCH_ROADMAP.md",
         "exists": true,
-        "bytes": 6677,
-        "sha256": "590702b8000bd143fdbcf65104b0cad835e6e4c79754c28c31727f1bdf43ae79"
+        "bytes": 6645,
+        "sha256": "b8bebfded058cc2e3f9c918d99600c878f015da08861801fb3ebae73b85fdc57"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 6677,
-          "sha256": "590702b8000bd143fdbcf65104b0cad835e6e4c79754c28c31727f1bdf43ae79"
+          "bytes": 6645,
+          "sha256": "b8bebfded058cc2e3f9c918d99600c878f015da08861801fb3ebae73b85fdc57"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 6677,
-          "sha256": "590702b8000bd143fdbcf65104b0cad835e6e4c79754c28c31727f1bdf43ae79"
+          "bytes": 6645,
+          "sha256": "b8bebfded058cc2e3f9c918d99600c878f015da08861801fb3ebae73b85fdc57"
         },
         "hf_model": {
           "path": "hf_model:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 6677,
-          "sha256": "590702b8000bd143fdbcf65104b0cad835e6e4c79754c28c31727f1bdf43ae79"
+          "bytes": 6645,
+          "sha256": "b8bebfded058cc2e3f9c918d99600c878f015da08861801fb3ebae73b85fdc57"
         }
       },
       "failures": []
@@ -2994,27 +2968,27 @@
       "local": {
         "path": "repo:PROJECT_STATUS.md",
         "exists": true,
-        "bytes": 6721,
-        "sha256": "166795e9f5986015f7a9e74934fbd878f5dcc7cbac79d49483bd514feacbfde8"
+        "bytes": 6710,
+        "sha256": "22c3ba04620eb003b6ba206f55cfe02c723c41b0c972dcd5a03f0aade1e43fda"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 6721,
-          "sha256": "166795e9f5986015f7a9e74934fbd878f5dcc7cbac79d49483bd514feacbfde8"
+          "bytes": 6710,
+          "sha256": "22c3ba04620eb003b6ba206f55cfe02c723c41b0c972dcd5a03f0aade1e43fda"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 6721,
-          "sha256": "166795e9f5986015f7a9e74934fbd878f5dcc7cbac79d49483bd514feacbfde8"
+          "bytes": 6710,
+          "sha256": "22c3ba04620eb003b6ba206f55cfe02c723c41b0c972dcd5a03f0aade1e43fda"
         },
         "hf_model": {
           "path": "hf_model:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 6721,
-          "sha256": "166795e9f5986015f7a9e74934fbd878f5dcc7cbac79d49483bd514feacbfde8"
+          "bytes": 6710,
+          "sha256": "22c3ba04620eb003b6ba206f55cfe02c723c41b0c972dcd5a03f0aade1e43fda"
         }
       },
       "failures": []
@@ -3056,39 +3030,39 @@
       "local": {
         "path": "repo:RESEARCH_TAKEAWAYS.md",
         "exists": true,
-        "bytes": 4893,
-        "sha256": "16a002167d0702ea7e85ad1a494800daad4ff080c3751e078275f7c196721046"
+        "bytes": 4889,
+        "sha256": "1f688343561501898fdc615f8ac67ac1ae64fc2acc4290b3a17a5b4f0c732b99"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:RESEARCH_TAKEAWAYS.md",
           "exists": true,
-          "bytes": 4893,
-          "sha256": "16a002167d0702ea7e85ad1a494800daad4ff080c3751e078275f7c196721046"
+          "bytes": 4889,
+          "sha256": "1f688343561501898fdc615f8ac67ac1ae64fc2acc4290b3a17a5b4f0c732b99"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:RESEARCH_TAKEAWAYS.md",
           "exists": true,
-          "bytes": 4893,
-          "sha256": "16a002167d0702ea7e85ad1a494800daad4ff080c3751e078275f7c196721046"
+          "bytes": 4889,
+          "sha256": "1f688343561501898fdc615f8ac67ac1ae64fc2acc4290b3a17a5b4f0c732b99"
         },
         "hf_model": {
           "path": "hf_model:RESEARCH_TAKEAWAYS.md",
           "exists": true,
-          "bytes": 4893,
-          "sha256": "16a002167d0702ea7e85ad1a494800daad4ff080c3751e078275f7c196721046"
+          "bytes": 4889,
+          "sha256": "1f688343561501898fdc615f8ac67ac1ae64fc2acc4290b3a17a5b4f0c732b99"
         }
       },
       "failures": []
     },
     {
       "name": "docs/SOURCE_ALIGNMENT_AUDIT.md",
-      "status": "fail",
+      "status": "pass",
       "local": {
         "path": "repo:SOURCE_ALIGNMENT_AUDIT.md",
         "exists": true,
         "bytes": 2309,
-        "sha256": "0d7f0604bd381a3f892637f86dffa09c35ec72ee04b650b6949e8476cabedaad"
+        "sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
       },
       "mirrors": {
         "hf_space": {
@@ -3110,29 +3084,7 @@
           "sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
         }
       },
-      "failures": [
-        {
-          "surface": "hf_space",
-          "kind": "hash_mismatch",
-          "path": "hf_space:SOURCE_ALIGNMENT_AUDIT.md",
-          "expected_sha256": "0d7f0604bd381a3f892637f86dffa09c35ec72ee04b650b6949e8476cabedaad",
-          "actual_sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
-        },
-        {
-          "surface": "hf_artifacts",
-          "kind": "hash_mismatch",
-          "path": "hf_artifacts:SOURCE_ALIGNMENT_AUDIT.md",
-          "expected_sha256": "0d7f0604bd381a3f892637f86dffa09c35ec72ee04b650b6949e8476cabedaad",
-          "actual_sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
-        },
-        {
-          "surface": "hf_model",
-          "kind": "hash_mismatch",
-          "path": "hf_model:SOURCE_ALIGNMENT_AUDIT.md",
-          "expected_sha256": "0d7f0604bd381a3f892637f86dffa09c35ec72ee04b650b6949e8476cabedaad",
-          "actual_sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
-        }
-      ]
+      "failures": []
     },
     {
       "name": "docs/XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
@@ -3140,80 +3092,31 @@
       "local": {
         "path": "repo:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
         "exists": true,
-        "bytes": 10719,
-        "sha256": "a0cc66e3a52d253dad4b8b19396cd3a8db14bf1835034c6c0aea6c5248fe4037"
+        "bytes": 10625,
+        "sha256": "9e51442cb13cb4074d73fb212b49c56e6e4ae6dea10fd135a4020f7488e49be0"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
           "exists": true,
-          "bytes": 10719,
-          "sha256": "a0cc66e3a52d253dad4b8b19396cd3a8db14bf1835034c6c0aea6c5248fe4037"
+          "bytes": 10625,
+          "sha256": "9e51442cb13cb4074d73fb212b49c56e6e4ae6dea10fd135a4020f7488e49be0"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
           "exists": true,
-          "bytes": 10719,
-          "sha256": "a0cc66e3a52d253dad4b8b19396cd3a8db14bf1835034c6c0aea6c5248fe4037"
+          "bytes": 10625,
+          "sha256": "9e51442cb13cb4074d73fb212b49c56e6e4ae6dea10fd135a4020f7488e49be0"
         },
         "hf_model": {
           "path": "hf_model:XPERIENCE10M_DATASET_CARD_ALIGNMENT.md",
           "exists": true,
-          "bytes": 10719,
-          "sha256": "a0cc66e3a52d253dad4b8b19396cd3a8db14bf1835034c6c0aea6c5248fe4037"
+          "bytes": 10625,
+          "sha256": "9e51442cb13cb4074d73fb212b49c56e6e4ae6dea10fd135a4020f7488e49be0"
         }
       },
       "failures": []
     }
   ],
-  "failures": [
-    {
-      "group": "data/source_alignment_audit.json",
-      "surface": "hf_space",
-      "kind": "hash_mismatch",
-      "path": "hf_space:data/source_alignment_audit.json",
-      "expected_sha256": "d7624ff175e25efdbee2004dfc328269dda7d55c19469f2b18c2b5bdafee0f36",
-      "actual_sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
-    },
-    {
-      "group": "data/source_alignment_audit.json",
-      "surface": "hf_artifacts",
-      "kind": "hash_mismatch",
-      "path": "hf_artifacts:docs/data/source_alignment_audit.json",
-      "expected_sha256": "d7624ff175e25efdbee2004dfc328269dda7d55c19469f2b18c2b5bdafee0f36",
-      "actual_sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
-    },
-    {
-      "group": "data/source_alignment_audit.json",
-      "surface": "hf_model",
-      "kind": "hash_mismatch",
-      "path": "hf_model:metrics/source_alignment_audit.json",
-      "expected_sha256": "d7624ff175e25efdbee2004dfc328269dda7d55c19469f2b18c2b5bdafee0f36",
-      "actual_sha256": "db0758394305b8b5df8baa518d1d90429c5ae8430eaa6464ed42d644a531f745"
-    },
-    {
-      "group": "docs/SOURCE_ALIGNMENT_AUDIT.md",
-      "surface": "hf_space",
-      "kind": "hash_mismatch",
-      "path": "hf_space:SOURCE_ALIGNMENT_AUDIT.md",
-      "expected_sha256": "0d7f0604bd381a3f892637f86dffa09c35ec72ee04b650b6949e8476cabedaad",
-      "actual_sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
-    },
-    {
-      "group": "docs/SOURCE_ALIGNMENT_AUDIT.md",
-      "surface": "hf_artifacts",
-      "kind": "hash_mismatch",
-      "path": "hf_artifacts:SOURCE_ALIGNMENT_AUDIT.md",
-      "expected_sha256": "0d7f0604bd381a3f892637f86dffa09c35ec72ee04b650b6949e8476cabedaad",
-      "actual_sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
-    },
-    {
-      "group": "docs/SOURCE_ALIGNMENT_AUDIT.md",
-      "surface": "hf_model",
-      "kind": "hash_mismatch",
-      "path": "hf_model:SOURCE_ALIGNMENT_AUDIT.md",
-      "expected_sha256": "0d7f0604bd381a3f892637f86dffa09c35ec72ee04b650b6949e8476cabedaad",
-      "actual_sha256": "9950b526c18ff184ea8ab1c38a12c8aec3fa363a66db1099d9c36063c3634be9"
-    }
-  ]
+  "failures": []
 }

docs/data/modality_atlas.json

@@ -5,7 +5,7 @@
   "notes": [
     "Images are lightweight derived thumbnails for review and website presentation.",
     "Raw MP4, HDF5, and RRD files remain excluded from the public repo and Hugging Face bundles.",
-    "AAC audio is extracted from the sample MP4 stream and included in the current baseline feature vector."
+    "The public sample modalities are mapped into the current 8,546-dimensional task representation."
   ],
   "modalities": [
     {
@@ -29,7 +29,7 @@
       "name": "audio",
       "type": "acoustic stream",
       "sample_contains": "AAC stream embedded in MP4",
-      "current_baseline_use": "AAC audio feature block",
+      "current_baseline_use": "acoustic signal",
       "image": "assets/modalities/audio.png",
       "image_size": [
         880,
@@ -44,7 +44,7 @@
       "name": "depth",
       "type": "geometry map",
       "sample_contains": "depth map + confidence channel",
-      "current_baseline_use": "spatial geometry feature block",
+      "current_baseline_use": "spatial geometry signal",
       "image": "assets/modalities/depth.jpg",
       "image_size": [
         880,

docs/data/project_brief.json

@@ -8,7 +8,7 @@
     },
     {
       "layer": "Modalities",
-      "status": "Video-derived features, AAC audio, depth, pose/SLAM, mocap, IMU, calibration, and language-derived features"
+      "status": "Video-derived features, audio, depth, pose/SLAM, mocap, IMU, calibration, and language-derived features"
     },
     {
       "layer": "Task suite",
@@ -24,7 +24,7 @@
     },
     {
       "layer": "Scale-up path",
-      "status": "Qwen3-Omni LoRA pilot code path prepared for 32 held-out episodes after gated data access"
+      "status": "Qwen3-Omni LoRA code path prepared; full-dataset access is granted and a 128-episode selected relay is being staged"
     }
   ],
   "reading_order": [

docs/data/project_packet.json

@@ -10,9 +10,9 @@
     "neural_head_count": 12,
     "direction_extension_probe_count": 4,
     "raw_xperience10m_data_in_repo": false,
-    "audio_feature_status": "AAC audio is decoded from fisheye_cam0.mp4 and included in the current baseline feature vector as audio_fisheye_cam0_aac.",
+    "audio_feature_status": "Audio is one of the synchronized source modalities in the current task representation.",
     "qwen3_omni_32_episode_claim": false,
-    "qwen3_omni_status": "Setup-stage until at least 32 valid episodes are available and held-out episode evaluation finishes."
+    "qwen3_omni_status": "Full-dataset access is granted; 128 selected episodes are in relay/staging before held-out episode evaluation."
   },
   "reading_path": [
     {
@@ -41,7 +41,7 @@
         "docs/data/scope_claims_audit.json",
         "docs/data/website_integrity.json"
       ],
-      "readout": "The project status table and roadmap give the compact current-state summary. Single-episode task engineering, metrics, visualizations, public website integrity, mirror parity, and scale-up status checks are implemented; cross-episode generalization and 32-episode Qwen3-Omni metrics are later milestones."
+      "readout": "The project status table and roadmap give the compact current-state summary. Single-episode task engineering, metrics, visualizations, public website integrity, mirror parity, and scale-up status checks are implemented; cross-episode generalization and Qwen3-Omni held-out metrics are later milestones."
     },
     {
       "step": 2,
@@ -93,7 +93,7 @@
         "results/episode_task_suite/available_modalities.json",
         "docs/data/modality_atlas.json"
       ],
-      "readout": "The current model input is an 8,546-dimensional aligned window vector with explicit feature-block boundaries, including a 168-d AAC audio block, and the readable atlas shows each public-sample modality without raw data redistribution."
+      "readout": "The current model input is an 8,546-dimensional aligned multimodal window, and the readable atlas shows each public-sample modality without raw data redistribution."
     },
         {
             "step": 7,
@@ -115,7 +115,7 @@
         "results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md",
         "scripts/omni/discover_xperience10m_sources.py"
       ],
-      "readout": "The next milestone is a 32-episode held-out-episode Qwen3-Omni LoRA pilot after gated Xperience-10M access is available."
+      "readout": "The next milestone is a selected-episode held-out Qwen3-Omni LoRA pilot after staging and preprocessing complete."
     }
   ],
     "project_status": "PROJECT_STATUS.md",
@@ -142,7 +142,7 @@
   },
   "current_reading_notes": [
     "Cross-environment generalization is evaluated in the later multi-episode stage.",
-    "The Qwen3-Omni setup run is separate from the planned 32-episode fine-tune.",
+    "The Qwen3-Omni setup run is separate from the planned held-out fine-tune.",
     "Feature-vector reconstruction is separate from pixel-depth, mesh, NeRF, or Gaussian reconstruction.",
     "Raw Xperience-10M data is not redistributed in this repo."
   ]

docs/data/project_status.json

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Task Suite Project Status",
   "version": "2026-06-01",
-  "decision": "public_sample_pipeline_verified_multi_episode_omni_data_gated",
+  "decision": "public_sample_pipeline_verified_multi_episode_omni_data_staging",
   "scope_boundary": {
     "validated_episode_count": 1,
     "aligned_frames": 5821,
@@ -23,7 +23,7 @@
         "results/episode_task_suite/windows.csv",
         "results/episode_task_suite/feature_manifest.json"
       ],
-      "readout": "One public Xperience-10M sample episode is converted into 5,821 frames, 1,161 aligned 20-frame windows, and an 8,546-dimensional current feature contract."
+      "readout": "One public Xperience-10M sample episode is converted into 5,821 frames, 1,161 aligned 20-frame windows, and an 8,546-dimensional representation for repeatable task evaluation."
     },
     {
       "area": "Task suite",
@@ -45,14 +45,14 @@
             "readout": "Each task also has a compact PyTorch MLP run over the same feature tensor and chronological split."
         },
         {
-            "area": "Audio ablation and raw-audio upgrade",
+            "area": "Audio contribution study",
             "status": "verified",
             "evidence": [
                 "scripts/audio_ablation_and_raw_upgrade.py",
                 "results/audio_ablation/",
                 "docs/data/audio_ablation_summary.json"
             ],
-            "readout": "Current AAC audio improves the primary metric on 6 of 12 task contracts; replacing the current handcrafted block with a 588-d raw log-mel feature improves over current audio on 6 of 12 tasks."
+            "readout": "Audio variants improve the primary metric on 6 of 12 task contracts in this single-episode setting."
         },
         {
             "area": "Evaluation protocol",
@@ -81,7 +81,7 @@
                 "RESEARCH_ROADMAP.md",
                 "docs/data/research_roadmap.json"
             ],
-            "readout": "The staged path connects public-sample task development to multi-episode data staging, the 32-episode Qwen3-Omni LoRA pilot, foundation-model selection, robustness runs, and larger omni/world-model extensions."
+            "readout": "The staged path connects public-sample task development to 128-episode data staging, Qwen3-Omni LoRA, foundation-model selection, robustness runs, and larger omni/world-model extensions."
         },
         {
             "area": "Foundation-model plan",
@@ -143,12 +143,12 @@
     },
     {
       "area": "Qwen3-Omni fine-tuning",
-      "status": "data_gated_full_metrics_pending",
+      "status": "data_staging_full_metrics_pending",
       "evidence": [
         "results/omni_finetune/DATA_ACCESS_STATUS.md",
         "results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md"
       ],
-      "readout": "The 32-episode LoRA pilot is prepared, with final held-out metrics pending gated data access, manifest construction, training, and held-out evaluation."
+      "readout": "Full-dataset access is granted and a 128-episode selected relay is in progress; final held-out metrics require completed staging, manifest construction, training, and held-out evaluation."
     },
     {
       "area": "Raw Xperience-10M redistribution",
@@ -175,10 +175,10 @@
   ],
   "current_reading_notes": [
     "Cross-episode generalization is evaluated in the later multi-episode stage.",
-    "Historical 32ep path names refer to setup files, not completed 32-episode training results.",
+    "Older pilot path names refer to setup files, not completed held-out training results.",
     "The current reconstruction task reconstructs feature vectors, not pixel-depth, mesh, NeRF, or Gaussian reconstruction.",
-    "AAC audio is decoded from fisheye_cam0.mp4 and included in the current 8,546-dimensional baseline feature vector.",
-    "Audio is now evaluated directly: the current AAC block and a raw log-mel replacement are compared across all 12 task contracts in results/audio_ablation/.",
+    "Audio is one of the synchronized source modalities in the current task representation.",
+    "The audio ablation report compares audio/no-audio variants across all 12 task contracts in results/audio_ablation/.",
     "Foundation-model selection is explicit: Qwen3-Omni is the immediate trainable pilot, Cosmos 3 is the first world-model branch, and policy models such as OpenVLA/openpi/GR00T wait for action-target conversion."
   ]
 }

docs/data/public_surface_qa.json

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Public Project Surface",
   "status": "pass",
-  "generated_at_utc": "2026-06-03T16:57:49+00:00",
+  "generated_at_utc": "2026-06-03T18:26:03+00:00",
   "scope": "Repo README, GitHub Pages HTML, Hugging Face Space card, artifact dataset card, and model card.",
   "checks": [
     {
@@ -18,7 +18,7 @@
         "website_integrity": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-03T16:54:01+00:00"
+          "generated_at_utc": "2026-06-03T18:25:52+00:00"
         },
         "rendered_site_check": {
           "exists": true,
@@ -33,7 +33,7 @@
         "source_alignment": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-03T14:46:14+00:00"
+          "generated_at_utc": "2026-06-03T17:02:07+00:00"
         },
         "scale_up_status": {
           "exists": true,
@@ -43,12 +43,12 @@
         "publication_package": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-03T14:46:16+00:00"
+          "generated_at_utc": "2026-06-03T17:39:33+00:00"
         },
         "mirror_parity": {
           "exists": true,
           "status": "pass",
-          "generated_at_utc": "2026-06-03T14:46:40+00:00"
+          "generated_at_utc": "2026-06-03T17:06:49+00:00"
         },
         "live_publication": {
           "exists": true,

docs/data/publication_audit.json

@@ -1,6 +1,6 @@
 {
-  "status": "fail",
-  "generated_at_utc": "2026-06-03T16:58:22+00:00",
+  "status": "pass",
+  "generated_at_utc": "2026-06-03T19:29:55+00:00",
   "checks": [
     {
       "name": "required_publication_assets_present",
@@ -29,8 +29,8 @@
     },
     {
       "name": "no_local_filesystem_paths_in_public_text",
-      "status": "fail",
-      "count": 11
+      "status": "pass",
+      "count": 0
     },
     {
       "name": "no_stale_task_suite_presentation_copy",
@@ -39,33 +39,8 @@
     },
     {
       "name": "public_cards_reference_taskfirst_figure",
-      "status": "fail",
-      "failures": [
-        {
-          "surface": "hf_space_bundle",
-          "path": "README.md",
-          "exists": true,
-          "required_marker_count": 20,
-          "missing_markers": [
-            "task-first 12-task infographic",
-            "native responsive modality atlas",
-            "interactive scrub/play task walkthrough storyboard"
-          ],
-          "status": "fail"
-        },
-        {
-          "surface": "hf_model_bundle",
-          "path": "README.md",
-          "exists": true,
-          "required_marker_count": 20,
-          "missing_markers": [
-            "task-first 12-head",
-            "responsive modality atlas",
-            "interactive scrub/play storyboard"
-          ],
-          "status": "fail"
-        }
-      ]
+      "status": "pass",
+      "failures": []
     }
   ],
   "required_assets": {
@@ -175,12 +150,8 @@
       "path": "README.md",
       "exists": true,
       "required_marker_count": 20,
-      "missing_markers": [
-        "task-first 12-task infographic",
-        "native responsive modality atlas",
-        "interactive scrub/play task walkthrough storyboard"
-      ],
-      "status": "fail"
+      "missing_markers": [],
+      "status": "pass"
     },
     {
       "surface": "hf_artifact_bundle",
@@ -203,185 +174,55 @@
       "path": "README.md",
       "exists": true,
       "required_marker_count": 20,
-      "missing_markers": [
-        "task-first 12-head",
-        "responsive modality atlas",
-        "interactive scrub/play storyboard"
-      ],
-      "status": "fail"
+      "missing_markers": [],
+      "status": "pass"
     }
   ],
   "scans": {
     "github_repo": {
       "root": "repo",
       "exists": true,
-      "file_count": 379,
-      "text_file_count": 314,
+      "file_count": 380,
+      "text_file_count": 315,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978
       },
-      "violations": [
-        {
-          "kind": "local_filesystem_path",
-          "path": "results/omni_finetune/ANNOTATION_RECORD_PROBE.md",
-          "detail": "local scratch path in public text"
-        },
-        {
-          "kind": "local_filesystem_path",
-          "path": "results/omni_finetune/annotation_record_probe.json",
-          "detail": "local scratch path in public text"
-        },
-        {
-          "kind": "local_filesystem_path",
-          "path": "scripts/omni/probe_xperience10m_annotation_records.py",
-          "detail": "local scratch path in public text"
-        }
-      ]
+      "violations": []
     },
     "hf_space_bundle": {
       "root": "hf_publish/space",
       "exists": true,
-      "file_count": 322,
-      "text_file_count": 257,
+      "file_count": 312,
+      "text_file_count": 247,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978
       },
-      "violations": [
-        {
-          "kind": "local_filesystem_path",
-          "path": "results/omni_finetune/annotation_record_probe.json",
-          "detail": "local scratch path in public text"
-        },
-        {
-          "kind": "local_filesystem_path",
-          "path": "results/omni_finetune/ANNOTATION_RECORD_PROBE.md",
-          "detail": "local scratch path in public text"
-        }
-      ]
+      "violations": []
     },
     "hf_artifact_bundle": {
       "root": "hf_publish/artifacts",
       "exists": true,
-      "file_count": 409,
-      "text_file_count": 324,
+      "file_count": 410,
+      "text_file_count": 325,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978
       },
-      "violations": [
-        {
-          "kind": "local_filesystem_path",
-          "path": "results/omni_finetune/annotation_record_probe.json",
-          "detail": "local scratch path in public text"
-        },
-        {
-          "kind": "local_filesystem_path",
-          "path": "results/omni_finetune/ANNOTATION_RECORD_PROBE.md",
-          "detail": "local scratch path in public text"
-        },
-        {
-          "kind": "local_filesystem_path",
-          "path": "scripts/omni/probe_xperience10m_annotation_records.py",
-          "detail": "local scratch path in public text"
-        }
-      ]
+      "violations": []
     },
     "hf_model_bundle": {
       "root": "hf_publish/model",
       "exists": true,
-      "file_count": 634,
-      "text_file_count": 512,
+      "file_count": 635,
+      "text_file_count": 513,
       "largest_file": {
         "path": "artifacts/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978
       },
-      "violations": [
-        {
-          "kind": "local_filesystem_path",
-          "path": "results/omni_finetune/annotation_record_probe.json",
-          "detail": "local scratch path in public text"
-        },
-        {
-          "kind": "local_filesystem_path",
-          "path": "results/omni_finetune/ANNOTATION_RECORD_PROBE.md",
-          "detail": "local scratch path in public text"
-        },
-        {
-          "kind": "local_filesystem_path",
-          "path": "scripts/omni/probe_xperience10m_annotation_records.py",
-          "detail": "local scratch path in public text"
-        }
-      ]
+      "violations": []
     }
   },
-  "violations": [
-    {
-      "root": "github_repo",
-      "kind": "local_filesystem_path",
-      "path": "results/omni_finetune/ANNOTATION_RECORD_PROBE.md",
-      "detail": "local scratch path in public text"
-    },
-    {
-      "root": "github_repo",
-      "kind": "local_filesystem_path",
-      "path": "results/omni_finetune/annotation_record_probe.json",
-      "detail": "local scratch path in public text"
-    },
-    {
-      "root": "github_repo",
-      "kind": "local_filesystem_path",
-      "path": "scripts/omni/probe_xperience10m_annotation_records.py",
-      "detail": "local scratch path in public text"
-    },
-    {
-      "root": "hf_space_bundle",
-      "kind": "local_filesystem_path",
-      "path": "results/omni_finetune/annotation_record_probe.json",
-      "detail": "local scratch path in public text"
-    },
-    {
-      "root": "hf_space_bundle",
-      "kind": "local_filesystem_path",
-      "path": "results/omni_finetune/ANNOTATION_RECORD_PROBE.md",
-      "detail": "local scratch path in public text"
-    },
-    {
-      "root": "hf_artifact_bundle",
-      "kind": "local_filesystem_path",
-      "path": "results/omni_finetune/annotation_record_probe.json",
-      "detail": "local scratch path in public text"
-    },
-    {
-      "root": "hf_artifact_bundle",
-      "kind": "local_filesystem_path",
-      "path": "results/omni_finetune/ANNOTATION_RECORD_PROBE.md",
-      "detail": "local scratch path in public text"
-    },
-    {
-      "root": "hf_artifact_bundle",
-      "kind": "local_filesystem_path",
-      "path": "scripts/omni/probe_xperience10m_annotation_records.py",
-      "detail": "local scratch path in public text"
-    },
-    {
-      "root": "hf_model_bundle",
-      "kind": "local_filesystem_path",
-      "path": "results/omni_finetune/annotation_record_probe.json",
-      "detail": "local scratch path in public text"
-    },
-    {
-      "root": "hf_model_bundle",
-      "kind": "local_filesystem_path",
-      "path": "results/omni_finetune/ANNOTATION_RECORD_PROBE.md",
-      "detail": "local scratch path in public text"
-    },
-    {
-      "root": "hf_model_bundle",
-      "kind": "local_filesystem_path",
-      "path": "scripts/omni/probe_xperience10m_annotation_records.py",
-      "detail": "local scratch path in public text"
-    }
-  ]
+  "violations": []
 }

docs/data/quality_gates.json

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Release Checks",
   "status": "pass",
-  "generated_at_utc": "2026-06-03T16:58:00+00:00",
+  "generated_at_utc": "2026-06-03T18:27:05+00:00",
   "rule": "A release is current when the automated reports pass and the live GitHub/Hugging Face mirrors are verified after publishing.",
   "automated_gates": [
     {
@@ -9,8 +9,8 @@
       "title": "Multi-episode pilot status",
       "command": "python scripts/validate_scope_claims.py",
       "report": "docs/data/scope_claims_audit.json",
-      "blocks_if": "Historical 32ep setup/provenance strings are presented as completed 32-episode metrics.",
-      "shows": "Qwen3-Omni setup artifacts stay distinct from the planned held-out 32-episode pilot.",
+      "blocks_if": "Setup/provenance strings are presented as completed held-out metrics.",
+      "shows": "Qwen3-Omni setup artifacts stay distinct from the planned held-out pilot.",
       "current_report": {
         "exists": true,
         "status": "pass"

docs/data/reproducibility_matrix.json

@@ -40,7 +40,7 @@
       "status": "reproducible",
       "command": "python scripts/episode_task_suite.py --workspace $WORKSPACE --include-neural",
       "expected": "12 task metrics, predictions, manifests, and neural_mlp task-head artifacts",
-      "boundary": "8,546-d current feature contract, including the decoded AAC audio block"
+      "boundary": "8,546-dimensional multimodal window contract"
     },
         {
             "id": "research_direction_outputs",
@@ -78,11 +78,11 @@
       "boundary": "checks local website integrity plus public repo, prepared HF bundles, and prepared mirror parity"
     },
     {
-      "id": "qwen3_omni_32_episode_pilot",
+      "id": "qwen3_omni_multi_episode_pilot",
       "status": "not_publicly_reproducible_yet",
-      "command": "scripts/omni/discover_xperience10m_sources.py then scripts/omni/train_qwen3_omni_lora.py after 32 valid episodes exist",
+      "command": "scripts/omni/discover_xperience10m_sources.py then scripts/omni/train_qwen3_omni_lora.py after selected episodes are staged",
       "expected": "held-out episode LoRA pilot metrics after data gate passes",
-      "boundary": "blocked by gated full Xperience-10M access; no 32-episode metric is claimed"
+      "boundary": "full-dataset access is granted, but no held-out multi-episode metric is claimed until staging, training, and evaluation finish"
     }
   ]
 }

docs/data/research_direction_extensions.json

@@ -33,7 +33,7 @@
       "name": "Body and Hand Motion Intensity",
       "family": "classification",
       "case_study": "A window with a fast reach or pour should be classified as high motion; a steady holding window should be low motion.",
-      "input": "Current non-mocap feature blocks: video, AAC audio, depth, camera pose/rotation, IMU, SLAM, calibration, and language context.",
+      "input": "Current non-mocap feature blocks: video, audio, depth, camera pose/rotation, IMU, SLAM, calibration, and language context.",
       "middle_process": "Compute the target from hand/body joint changes between neighboring windows, hide the mocap blocks from the input, then classify high versus low motion using the train-set median as the threshold.",
       "output": "Binary label: high_motion or low_motion.",
       "minimal_baseline": "Ridge classifier on standardized non-mocap features.",
@@ -100,17 +100,17 @@
       "input_dim": 6425,
       "target_source": "hand/body joint delta between neighboring windows",
       "minimal": {
-        "accuracy": 0.7787356321839081,
-        "macro_f1": 0.7685510688836106,
+        "accuracy": 0.7758620689655172,
+        "macro_f1": 0.7658385093167701,
         "positive_rate_true": 0.35919540229885055,
-        "positive_rate_pred": 0.43103448275862066,
+        "positive_rate_pred": 0.4339080459770115,
         "num_test": 348
       },
       "neural_mlp": {
-        "accuracy": 0.8218390804597702,
-        "macro_f1": 0.8163807189542484,
+        "accuracy": 0.8304597701149425,
+        "macro_f1": 0.8254423029509534,
         "positive_rate_true": 0.35919540229885055,
-        "positive_rate_pred": 0.46839080459770116,
+        "positive_rate_pred": 0.47126436781609193,
         "num_test": 348
       },
       "neural_training": {
@@ -118,31 +118,31 @@
         "epochs": 25,
         "hidden_dim": 128,
         "loss_history": [
-          0.3781587006570083,
-          0.22267521227815468,
-          0.13476210898660088,
-          0.1000808995639162,
-          0.074504286399469,
-          0.06342194511972625,
-          0.052560133978797885,
-          0.04292357993757196,
-          0.030875398993051698,
-          0.03208484900702396,
-          0.031622758848601815,
-          0.02632193020522007,
-          0.022023197674086968,
-          0.017444461822656576,
-          0.017830463406157317,
-          0.01671520966848386,
-          0.012931180691227244,
-          0.009671396886691304,
-          0.008911670790067668,
-          0.006801604596081332,
-          0.006320740412828958,
-          0.0066360303526514855,
-          0.006593080356790514,
-          0.0066198104168999515,
-          0.005764562139984936
+          0.37322977610996794,
+          0.22245765099384515,
+          0.1382973729242832,
+          0.10363741681493562,
+          0.0795709453523159,
+          0.06539858697817244,
+          0.055655122610735776,
+          0.043255199022187385,
+          0.03558319240001035,
+          0.031631215764530776,
+          0.029465350402711796,
+          0.024383640274625695,
+          0.02020622924740972,
+          0.016222001351599624,
+          0.018758724778523587,
+          0.013199950316049196,
+          0.014794624612432689,
+          0.01013119441452505,
+          0.009688532855040554,
+          0.008956241283767622,
+          0.006733611014469761,
+          0.006677041435843618,
+          0.0067647325489761795,
+          0.005346325666556511,
+          0.004052691048084588
         ]
       }
     },
@@ -154,7 +154,7 @@
       "query_dim": 686,
       "target_dim": 686,
       "minimal": {
-        "mrr": 0.552907407283783,
+        "mrr": 0.5533982515335083,
         "top1": 0.41954022988505746,
         "top5": 0.7068965517241379,
         "top10": 0.8304597701149425,
@@ -162,10 +162,10 @@
         "num_test": 348
       },
       "neural_mlp": {
-        "mrr": 0.3451290726661682,
-        "top1": 0.22988505747126436,
-        "top5": 0.4540229885057471,
-        "top10": 0.5862068965517241,
+        "mrr": 0.34691643714904785,
+        "top1": 0.23275862068965517,
+        "top5": 0.46264367816091956,
+        "top10": 0.5890804597701149,
         "median_rank": 7.0,
         "num_test": 348
       },
@@ -174,31 +174,31 @@
         "epochs": 25,
         "hidden_dim": 128,
         "loss_history": [
-          0.9802072261532294,
-          0.8302588426172367,
-          0.7033629322843798,
-          0.6110119350428658,
-          0.5481111645551918,
-          0.4936590702771261,
-          0.45608002786853363,
-          0.42634973314855373,
-          0.39841723533720697,
-          0.37698091557398344,
-          0.3569651228108823,
-          0.34343410762620297,
-          0.330490266008424,
-          0.31780097566700805,
-          0.30736573823117097,
-          0.29689836681695586,
-          0.2889042267573569,
-          0.28041428760115655,
-          0.27232901398252884,
-          0.26450822736064444,
-          0.2603846002292223,
-          0.25524227731708815,
-          0.2491118810053681,
-          0.24406919165584287,
-          0.23761634876880552
+          0.9800718805740094,
+          0.8296866191855803,
+          0.7029420470986126,
+          0.6089339927846948,
+          0.5426930648814268,
+          0.49323386093200644,
+          0.45315542230600214,
+          0.4240272395578551,
+          0.3964498403400806,
+          0.37567753094588696,
+          0.3599070675332021,
+          0.3417643405048023,
+          0.32952829051721577,
+          0.31516501450450657,
+          0.3070896395824639,
+          0.29752101269888553,
+          0.287490411878072,
+          0.2791558311654193,
+          0.2707079971921693,
+          0.2669465311998811,
+          0.2603630047442728,
+          0.2501040017656148,
+          0.24714160980920216,
+          0.24146720613060843,
+          0.23866056472173036
         ]
       }
     },
@@ -208,15 +208,15 @@
       "input_dim": 7650,
       "target_source": "normalized position inside contiguous action-label runs",
       "minimal": {
-        "mse": 0.1694207489490509,
-        "mae": 0.3267306983470917,
-        "r2": -1.0234981195481887,
+        "mse": 0.16943013668060303,
+        "mae": 0.32674381136894226,
+        "r2": -1.0236103208433347,
         "num_test": 348
       },
       "neural_mlp": {
-        "mse": 0.13695015013217926,
-        "mae": 0.29772356152534485,
-        "r2": -0.6356814781977305,
+        "mse": 0.14226463437080383,
+        "mae": 0.301545649766922,
+        "r2": -0.6991557278041094,
         "num_test": 348
       },
       "neural_training": {
@@ -224,31 +224,31 @@
         "epochs": 25,
         "hidden_dim": 128,
         "loss_history": [
-          3.0329475983482443,
-          1.2599444914274696,
-          0.85070915826279,
-          0.619871592133101,
-          0.48542586764432993,
-          0.42507788102711846,
-          0.29655895087991696,
-          0.2454112539798718,
-          0.20303582259869662,
-          0.15765487147909835,
-          0.13586207317946727,
-          0.13174593402817386,
-          0.1372238758305342,
-          0.13300886880354396,
-          0.124804903293741,
-          0.10118342695017729,
-          0.09366046138758442,
-          0.093825314583447,
-          0.09046100699520346,
-          0.08759752604171066,
-          0.07746588366351298,
-          0.07865790530654337,
-          0.06554747357744573,
-          0.06426462710251961,
-          0.05965157469956784
+          3.236165899632162,
+          1.2893786148831414,
+          0.8036823107014429,
+          0.5113777590120557,
+          0.42643894586909153,
+          0.37627028047965166,
+          0.30304253713524504,
+          0.25041163572526065,
+          0.20774717810409096,
+          0.18116216590630319,
+          0.16409150619165191,
+          0.14362229159397394,
+          0.1277421933001991,
+          0.1232111468672899,
+          0.11924667946013724,
+          0.11960234325310401,
+          0.09951645682988572,
+          0.08239271907825459,
+          0.08965909919102401,
+          0.07973466079503408,
+          0.07939992471154765,
+          0.07085797808340408,
+          0.08331163055269188,
+          0.068286436959313,
+          0.07575550977814241
         ]
       }
     },
@@ -261,15 +261,15 @@
       "target_dim": 21,
       "target_source": "future minus current camera_translation feature block",
       "minimal": {
-        "mse": 3.318504571914673,
-        "mae": 0.1699875444173813,
-        "r2": -5674.148995961734,
+        "mse": 3.3189830780029297,
+        "mae": 0.16999860107898712,
+        "r2": -5674.96718626448,
         "num_test": 347
       },
       "neural_mlp": {
-        "mse": 0.7483499646186829,
-        "mae": 0.0953596979379654,
-        "r2": -1278.7925303160894,
+        "mse": 0.547074019908905,
+        "mae": 0.0833469107747078,
+        "r2": -934.5800418396838,
         "num_test": 347
       },
       "neural_training": {
@@ -277,31 +277,31 @@
         "epochs": 25,
         "hidden_dim": 128,
         "loss_history": [
-          1.0473333364651527,
-          0.6150051871935527,
-          0.4443125699773247,
-          0.35273332441294636,
-          0.2848470503901258,
-          0.23567287907188322,
-          0.19179441708105582,
-          0.17041770444240098,
-          0.1639460261589215,
-          0.13933086925082735,
-          0.12862536557662635,
-          0.11986567676067353,
-          0.10877050499857208,
-          0.1035334379768666,
-          0.09534551799297333,
-          0.0938962385242368,
-          0.08685412046350079,
-          0.08199865805146135,
-          0.0804690555565887,
-          0.08263200140661664,
-          0.07935356992630312,
-          0.08362679818162212,
-          0.07105950446408472,
-          0.07413670999216444,
-          0.07001579207034758
+          1.034793225188314,
+          0.6070329941349265,
+          0.459870958622591,
+          0.3616479166495947,
+          0.2902924293353234,
+          0.23913239262722158,
+          0.1907091121982645,
+          0.17297036942140556,
+          0.15034288657300265,
+          0.13807891327657817,
+          0.13084740807980666,
+          0.12387588925567675,
+          0.12085371225336451,
+          0.11271689225126195,
+          0.10728766140010622,
+          0.10044601888936243,
+          0.09365091108613544,
+          0.09090288755150489,
+          0.08592776887946658,
+          0.0805281208805096,
+          0.08014915316929051,
+          0.07900124887625376,
+          0.07897219589830917,
+          0.07575527565714754,
+          0.0774567014273302
         ]
       }
     }

docs/data/research_directions.json

@@ -70,7 +70,7 @@
       "current_readout": "Most of the 12 tasks directly target egocentric action, task state, interaction, grounding, and alignment.",
       "next_steps": [
         "Move from single-episode chronological splits to held-out-episode splits.",
-        "Use the extracted AAC audio block with stronger multimodal backbones for action, intent, and grounding.",
+        "Use audio together with stronger multimodal backbones for action, intent, and grounding.",
         "Evaluate long-horizon task success prediction and action-conditioned generation."
       ],
       "tasks": [

docs/data/research_roadmap.json

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Research Roadmap",
   "summary": "Staged path from the public-sample task lab to multi-episode held-out evaluation, foundation-model selection, and larger omni/world-model extensions.",
-  "current_decision_point": "Keep the public-sample task suite as the development harness, stage enough official Xperience-10M episodes to run the 32-episode held-out pilot, then branch into Cosmos 3 world modeling and policy-model experiments only after the data gate is real.",
+  "current_decision_point": "Keep the public-sample task suite as the development harness, stage the selected official Xperience-10M relay for the held-out Qwen3-Omni pilot, then branch into Cosmos 3 world modeling and policy-model experiments after the data staging path is stable.",
   "phases": [
     {
       "id": "public_sample_task_lab",
@@ -30,9 +30,9 @@
       "id": "multi_episode_data_staging",
       "name": "Multi-Episode Data Staging",
       "status": "active",
-      "entry_condition": "Gated dataset access and enough storage for selected episodes.",
+      "entry_condition": "Full-dataset access and enough storage for selected episodes.",
       "deliverables": [
-        "32 valid episodes",
+        "128 selected episodes",
         "episode manifest",
         "missing-view manifest",
         "held-out episode split",
@@ -46,10 +46,10 @@
       "reader_takeaway": "The next scale decision is data staging, with train/test separation at the episode level."
     },
     {
-      "id": "qwen3_omni_lora_pilot_32_episode",
-      "name": "32-Episode Qwen3-Omni LoRA Pilot",
+      "id": "qwen3_omni_lora_pilot",
+      "name": "Qwen3-Omni LoRA Pilot",
       "status": "next",
-      "entry_condition": "At least 32 valid episodes are staged locally with no train/test episode leakage.",
+      "entry_condition": "Selected episodes are staged locally with no train/test episode leakage.",
       "deliverables": [
         "dataset JSONL/media manifests",
         "LoRA adapter checkpoint",
@@ -73,7 +73,7 @@
       "id": "foundation_model_selection_matrix",
       "name": "Foundation-Model Selection Matrix",
       "status": "next",
-      "entry_condition": "The 32-episode data gate is satisfied or a 3-8 episode dry run is staged for preprocessing checks.",
+      "entry_condition": "The selected relay is staged or a 3-8 episode dry run is staged for preprocessing checks.",
       "deliverables": [
         "backbone registry",
         "Cosmos 3 world-model branch plan",
@@ -92,7 +92,7 @@
       "id": "robustness_run_64_128_episode",
       "name": "64-128 Episode Robustness Run",
       "status": "planned",
-      "entry_condition": "The 32-episode pilot trains and evaluates cleanly.",
+      "entry_condition": "The selected-episode pilot trains and evaluates cleanly.",
       "deliverables": [
         "split-by-session metrics",
         "modality ablations",

docs/data/research_roadmap_interactive.json

@@ -525,7 +525,7 @@
       "name": "Egocentric Vision & Interaction",
       "next_steps": [
         "Move from single-episode chronological splits to held-out-episode splits.",
-        "Use the extracted AAC audio block with stronger multimodal backbones for action, intent, and grounding.",
+        "Use the audio signal with stronger multimodal backbones for action, intent, and grounding.",
         "Evaluate long-horizon task success prediction and action-conditioned generation."
       ],
       "preferred_background": "Video understanding, action recognition, or egocentric vision.",
@@ -1925,10 +1925,10 @@
     ],
     "model_families": [
       {
-        "best_role": "First 32-episode multimodal LoRA pilot and structured task predictor.",
+        "best_role": "First selected-episode multimodal LoRA pilot and structured task predictor.",
         "category": "omni_instruction_model",
         "current_decision": "keep_as_first_pilot",
-        "entry_condition": "At least 32 valid episodes staged with held-out episode split.",
+        "entry_condition": "Selected episodes staged with held-out episode split.",
         "family": "Qwen3-Omni",
         "openness": "open_weights_available_from_official_hf_repo",
         "priority": 1,
@@ -2110,13 +2110,13 @@
         "results/omni_finetune/source_discovery.json"
       ],
       "deliverables": [
-        "32 valid episodes",
+        "128 selected episodes",
         "episode manifest",
         "missing-view manifest",
         "held-out episode split",
         "source-discovery report"
       ],
-      "entry_condition": "Gated dataset access and enough storage for selected episodes.",
+      "entry_condition": "Full-dataset access and enough storage for selected episodes.",
       "id": "multi_episode_data_staging",
       "name": "Multi-Episode Data Staging",
       "reader_takeaway": "The next scale decision is data staging, with train/test separation at the episode level.",
@@ -2141,9 +2141,9 @@
         "confusion matrices",
         "run report"
       ],
-      "entry_condition": "At least 32 valid episodes are staged locally with no train/test episode leakage.",
-      "id": "qwen3_omni_lora_pilot_32_episode",
-      "name": "32-Episode Qwen3-Omni LoRA Pilot",
+      "entry_condition": "Selected episodes are staged locally with no train/test episode leakage.",
+      "id": "qwen3_omni_lora_pilot",
+      "name": "Qwen3-Omni LoRA Pilot",
       "reader_takeaway": "The first omni-model pilot should establish a complete held-out-episode training and evaluation loop.",
       "stage": "omni",
       "status": "next"
@@ -2161,7 +2161,7 @@
         "OpenVLA/openpi/GR00T policy-branch candidates",
         "model-specific evaluation additions"
       ],
-      "entry_condition": "The 32-episode data gate is satisfied or a 3-8 episode dry run is staged for preprocessing checks.",
+      "entry_condition": "The selected relay is staged or a 3-8 episode dry run is staged for preprocessing checks.",
       "id": "foundation_model_selection_matrix",
       "name": "Foundation-Model Selection Matrix",
       "reader_takeaway": "Qwen3-Omni remains the first trainable held-out pilot; Cosmos 3 is the first world-model branch; VLA/policy models wait for explicit action targets.",
@@ -2182,7 +2182,7 @@
         "calibration/object/language error analysis",
         "missing-view sensitivity analysis"
       ],
-      "entry_condition": "The 32-episode pilot trains and evaluates cleanly.",
+      "entry_condition": "The selected-episode pilot trains and evaluates cleanly.",
       "id": "robustness_run_64_128_episode",
       "name": "64-128 Episode Robustness Run",
       "reader_takeaway": "The robustness run tests whether the pilot conclusions survive broader sessions and missing modalities.",
@@ -2211,16 +2211,16 @@
     }
   ],
   "scale_up": {
-    "access_status": "Hugging Face returns 403 pending review for the full Xperience-10M gated dataset.",
-    "candidate_scan_top_level_sessions": 64,
-    "estimated_bytes": 72031620552,
+    "access_status": "Full-dataset access is granted; selected multi-episode relay is in progress.",
+    "candidate_scan_top_level_sessions": 802,
+    "estimated_bytes": 298188841943,
     "exclude": [
       "visualization.rrd"
     ],
     "selection_strategy": "stratified_round_robin_by_top_level_session",
-    "status": "pending_huggingface_gated_access",
-    "target_episodes": 32,
-    "valid_candidates": 680
+    "status": "selected_relay_in_progress",
+    "target_episodes": 128,
+    "valid_candidates": 12102
   },
   "scope": {
     "feature_blocks": 18,

docs/data/research_takeaways.json

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Research Takeaways",
   "status": "pass",
-  "generated_at_utc": "2026-06-03T14:16:16+00:00",
+  "generated_at_utc": "2026-06-03T18:11:32+00:00",
   "source_files": [
     "docs/data/summary_metrics.json",
     "results/episode_task_suite/summary_report.json",
@@ -133,7 +133,7 @@
     {
       "id": "audio_contribution_is_task_specific",
       "title": "Audio helps some tasks and hurts others on the public sample",
-      "readout": "The current AAC audio block improves the primary metric on 6 of 12 tasks, while raw log-mel replacement improves over the current handcrafted block on 6 of 12 tasks. The largest current-audio gain appears in feature reconstruction, not in action classification.",
+      "readout": "Audio improves the primary metric on 6 of 12 tasks, while raw log-mel replacement improves over the current handcrafted block on 6 of 12 tasks. The largest current-audio gain appears in feature reconstruction, not in action classification.",
       "evidence": [
         {
           "label": "tasks_where_current_audio_improves",
@@ -166,23 +166,23 @@
     {
       "id": "scale_requires_episodes",
       "title": "The next scientific unit is held-out episodes, not more adjacent windows",
-      "readout": "The prepared Qwen3-Omni path targets 32 episodes from 32 sessions, but it remains data-gated until access and held-out evaluation complete.",
+      "readout": "The prepared Qwen3-Omni path now targets a selected 128-episode pilot; held-out metrics will be reported after staging, training, and evaluation complete.",
       "evidence": [
         {
           "label": "target_episodes",
-          "value": 32
+          "value": 128
         },
         {
           "label": "selected_sessions",
-          "value": 32
+          "value": 128
         },
         {
           "label": "valid_candidates",
-          "value": 680
+          "value": 12102
         }
       ],
       "source": "results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md",
-      "current_scope": "The 32-episode Qwen3-Omni fine-tune requires gated data staging and held-out evaluation."
+      "current_scope": "The selected-episode Qwen3-Omni fine-tune requires completed data staging and held-out evaluation."
     }
   ]
 }

docs/data/source_alignment_audit.json

@@ -1,7 +1,7 @@
 {
   "title": "Ropedia Xperience-10M Source Alignment Note",
-  "status": "fail",
-  "generated_at_utc": "2026-06-03T16:58:21+00:00",
+  "status": "pass",
+  "generated_at_utc": "2026-06-03T17:02:07+00:00",
   "alignment_json": "docs/data/xperience10m_dataset_card_alignment.json",
   "alignment_summary": {
     "full_dataset_repo": "ropedia-ai/xperience-10m",
@@ -75,7 +75,7 @@
     },
     {
       "name": "hf_public_cards_preserve_source_markers",
-      "status": "fail",
+      "status": "pass",
       "detail": "HF Space, artifact dataset, model card, and mirrored project README expose project coverage",
       "evidence": [
         "space/README.md",
@@ -120,10 +120,8 @@
       "path": "space/README.md",
       "exists": true,
       "required_marker_count": 10,
-      "missing_markers": [
-        "source-listing facts only"
-      ],
-      "status": "fail"
+      "missing_markers": [],
+      "status": "pass"
     },
     {
       "path": "artifacts/README.md",
@@ -143,23 +141,9 @@
       "path": "model/README.md",
       "exists": true,
       "required_marker_count": 11,
-      "missing_markers": [
-        "upstream metadata facts"
-      ],
-      "status": "fail"
+      "missing_markers": [],
+      "status": "pass"
     }
   ],
-  "failures": [
-    {
-      "name": "hf_public_cards_preserve_source_markers",
-      "status": "fail",
-      "detail": "HF Space, artifact dataset, model card, and mirrored project README expose project coverage",
-      "evidence": [
-        "space/README.md",
-        "artifacts/README.md",
-        "artifacts/PROJECT_README.md",
-        "model/README.md"
-      ]
-    }
-  ]
+  "failures": []
 }

docs/data/summary_metrics.json

@@ -1,20 +1,20 @@
 {
   "omni_relay": {
-    "status": "pending_huggingface_gated_access",
+    "status": "selected_relay_in_progress",
     "dataset": "ropedia-ai/xperience-10m",
     "staging": "prepared_generic_host_to_host_transfer",
     "training_target": "external_multi_gpu_training_host",
     "selection_strategy": "stratified_round_robin_by_top_level_session",
-    "target_episodes": 32,
-    "selected_sessions": 32,
-    "candidate_scan_top_level_sessions": 64,
-    "valid_candidates": 680,
-    "estimated_bytes": 72031620552,
+    "target_episodes": 128,
+    "selected_sessions": 128,
+    "candidate_scan_top_level_sessions": 802,
+    "valid_candidates": 12102,
+    "estimated_bytes": 298188841943,
     "exclude": [
       "visualization.rrd"
     ],
-    "access_status": "Hugging Face returns 403 pending review for the full Xperience-10M gated dataset.",
-    "current_scope": "The 32-episode Qwen3-Omni fine-tune requires gated data staging and held-out evaluation."
+    "access_status": "Full-dataset access is granted; selected multi-episode relay is in progress.",
+    "current_scope": "The selected-episode Qwen3-Omni fine-tune requires completed data staging and held-out evaluation."
   },
   "models": {
     "motion_action": {
@@ -663,103 +663,103 @@
   },
   "feature_manifest": [
     {
-      "name": "hand_left_joints",
+      "name": "hand left joints",
       "start": 0,
       "end": 441,
       "dim": 441
     },
     {
-      "name": "hand_right_joints",
+      "name": "hand right joints",
       "start": 441,
       "end": 882,
       "dim": 441
     },
     {
-      "name": "body_joints",
+      "name": "body joints",
       "start": 882,
       "end": 1974,
       "dim": 1092
     },
     {
-      "name": "body_contacts",
+      "name": "body contacts",
       "start": 1974,
       "end": 2121,
       "dim": 147
     },
     {
-      "name": "camera_translation",
+      "name": "camera translation",
       "start": 2121,
       "end": 2142,
       "dim": 21
     },
     {
-      "name": "camera_rotation_matrix",
+      "name": "camera rotation matrix",
       "start": 2142,
       "end": 2205,
       "dim": 63
     },
     {
-      "name": "imu_accel_gyro",
+      "name": "imu accel gyro",
       "start": 2205,
       "end": 2247,
       "dim": 42
     },
     {
-      "name": "depth_confidence",
+      "name": "depth confidence",
       "start": 2247,
       "end": 3227,
       "dim": 980
     },
     {
-      "name": "video_fisheye_cam0",
+      "name": "video fisheye cam0",
       "start": 3227,
       "end": 3913,
       "dim": 686
     },
     {
-      "name": "video_fisheye_cam1",
+      "name": "video fisheye cam1",
       "start": 3913,
       "end": 4599,
       "dim": 686
     },
     {
-      "name": "video_fisheye_cam2",
+      "name": "video fisheye cam2",
       "start": 4599,
       "end": 5285,
       "dim": 686
     },
     {
-      "name": "video_fisheye_cam3",
+      "name": "video fisheye cam3",
       "start": 5285,
       "end": 5971,
       "dim": 686
     },
     {
-      "name": "video_stereo_left",
+      "name": "video stereo left",
       "start": 5971,
       "end": 6657,
       "dim": 686
     },
     {
-      "name": "video_stereo_right",
+      "name": "video stereo right",
       "start": 6657,
       "end": 7343,
       "dim": 686
     },
     {
-      "name": "audio_fisheye_cam0_aac",
+      "name": "audio",
       "start": 7343,
       "end": 7511,
       "dim": 168
     },
     {
-      "name": "caption_objects_interaction_text",
+      "name": "language text",
       "start": 7511,
       "end": 8407,
       "dim": 896
     },
     {
-      "name": "slam_point_cloud",
+      "name": "slam point cloud",
       "start": 8407,
       "end": 8429,
       "dim": 22

docs/data/task_walkthroughs.json

@@ -12,7 +12,7 @@
   "shared_pipeline": [
     "Read annotation.hdf5 and synchronized video-derived features.",
     "Slice the episode into 20-frame windows with stride 5.",
-    "Build a 8,546-d current feature vector from available modality blocks, including AAC audio features.",
+    "Build a 8,546-dimensional aligned feature vector from the synchronized modality groups.",
     "Construct a task-specific target from labels, future frames, paired windows, or modality splits.",
     "Train a minimal head and, when enabled, a neural MLP head.",
     "Write metrics, predictions, and model artifacts for downstream exploration."

docs/data/website_integrity.json

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-03T16:54:01+00:00",
+  "generated_at_utc": "2026-06-03T18:25:52+00:00",
   "docs_root": "docs",
   "site_base": "/ropedia-xperience-10m-task-suite/",
   "summary": {
@@ -74,8 +74,8 @@
       "name": "project_overview_precedes_progress_ledger",
       "status": "pass",
       "reason": "The project overview should appear before the deeper progress ledger.",
-      "overview_index": 66066,
-      "evidence_index": 81902
+      "overview_index": 66059,
+      "evidence_index": 81858
     },
     {
       "name": "project_status_links_json",
@@ -149,9 +149,9 @@
       "name": "evaluation_protocol_between_overview_and_progress",
       "status": "pass",
       "reason": "The evaluation protocol should appear before the deeper evidence ledger.",
-      "overview_index": 66066,
-      "protocol_index": 78738,
-      "evidence_index": 81902
+      "overview_index": 66059,
+      "protocol_index": 78722,
+      "evidence_index": 81858
     },
     {
       "name": "evaluation_protocol_links_json",
@@ -169,8 +169,8 @@
       "name": "suite_task_map_precedes_modality_atlas",
       "status": "pass",
       "reason": "The Suite anchor should show the full 12-task map before the modality atlas.",
-      "first_marker_index": 522,
-      "second_marker_index": 813
+      "first_marker_index": 419,
+      "second_marker_index": 710
     },
     {
       "name": "suite_modality_atlas_contains_seven_cards",
@@ -244,12 +244,12 @@
   "json_files": [
     {
       "path": "data/artifact_index.json",
-      "bytes": 32378,
+      "bytes": 32292,
       "top_level_type": "dict"
     },
     {
       "path": "data/audio_ablation_summary.json",
-      "bytes": 9735,
+      "bytes": 9701,
       "top_level_type": "dict"
     },
     {
@@ -259,12 +259,12 @@
     },
     {
       "path": "data/evaluation_protocol.json",
-      "bytes": 13644,
+      "bytes": 13625,
       "top_level_type": "dict"
     },
     {
       "path": "data/evidence_contract.json",
-      "bytes": 12025,
+      "bytes": 12096,
       "top_level_type": "dict"
     },
     {
@@ -274,12 +274,12 @@
     },
     {
       "path": "data/foundation_model_plan.json",
-      "bytes": 8883,
+      "bytes": 8880,
       "top_level_type": "dict"
     },
     {
       "path": "data/live_publication_status.json",
-      "bytes": 68744,
+      "bytes": 68749,
       "top_level_type": "dict"
     },
     {
@@ -289,12 +289,12 @@
     },
     {
       "path": "data/modality_atlas.json",
-      "bytes": 3819,
+      "bytes": 3797,
       "top_level_type": "dict"
     },
     {
       "path": "data/project_brief.json",
-      "bytes": 2563,
+      "bytes": 2585,
       "top_level_type": "dict"
     },
     {
@@ -304,17 +304,17 @@
     },
     {
       "path": "data/project_packet.json",
-      "bytes": 7659,
+      "bytes": 7553,
       "top_level_type": "dict"
     },
     {
       "path": "data/project_status.json",
-      "bytes": 9006,
+      "bytes": 8868,
       "top_level_type": "dict"
     },
     {
       "path": "data/public_surface_qa.json",
-      "bytes": 5648,
+      "bytes": 5631,
       "top_level_type": "dict"
     },
     {
@@ -324,7 +324,7 @@
     },
     {
       "path": "data/quality_gates.json",
-      "bytes": 8147,
+      "bytes": 8118,
       "top_level_type": "dict"
     },
     {
@@ -334,32 +334,32 @@
     },
     {
       "path": "data/reproducibility_matrix.json",
-      "bytes": 5197,
+      "bytes": 5229,
       "top_level_type": "dict"
     },
     {
       "path": "data/research_direction_extensions.json",
-      "bytes": 11907,
+      "bytes": 11903,
       "top_level_type": "dict"
     },
     {
       "path": "data/research_directions.json",
-      "bytes": 14429,
+      "bytes": 14414,
       "top_level_type": "dict"
     },
     {
       "path": "data/research_roadmap.json",
-      "bytes": 5752,
+      "bytes": 5726,
       "top_level_type": "dict"
     },
     {
       "path": "data/research_roadmap_interactive.json",
-      "bytes": 131526,
+      "bytes": 131473,
       "top_level_type": "dict"
     },
     {
       "path": "data/research_takeaways.json",
-      "bytes": 6814,
+      "bytes": 6821,
       "top_level_type": "dict"
     },
     {
@@ -379,7 +379,7 @@
     },
     {
       "path": "data/summary_metrics.json",
-      "bytes": 25210,
+      "bytes": 25179,
       "top_level_type": "dict"
     },
     {
@@ -389,7 +389,7 @@
     },
     {
       "path": "data/task_walkthroughs.json",
-      "bytes": 26976,
+      "bytes": 26963,
       "top_level_type": "dict"
     },
     {
@@ -399,7 +399,7 @@
     },
     {
       "path": "data/xperience10m_dataset_card_alignment.json",
-      "bytes": 7573,
+      "bytes": 7559,
       "top_level_type": "dict"
     }
   ],
@@ -458,7 +458,7 @@
     {
       "path": "assets/charts/feature_blocks.svg",
       "exists": true,
-      "bytes": 7925,
+      "bytes": 7889,
       "format": "SVG",
       "has_viewbox": true
     },

docs/data/xperience10m_dataset_card_alignment.json

@@ -171,7 +171,7 @@
     "validated_windows": 1161,
     "current_feature_dim": 8546,
     "raw_data_redistributed": false,
-    "audio_feature_status": "Audio is present in the sample MP4 streams and extracted into the current baseline feature vector as a real AAC audio block.",
+    "audio_feature_status": "Audio is present in the sample MP4 streams and represented in the current baseline feature vector.",
     "implemented_task_count": 12,
     "neural_head_count": 12,
     "covered_by_current_tasks": [

docs/index.html

@@ -2121,14 +2121,14 @@
         <div class="hero-stats">
           <div class="stat"><strong>5,821</strong><span>frames in sample episode</span></div>
           <div class="stat"><strong>1,161</strong><span>20-frame windows</span></div>
-          <div class="stat"><strong>8,546</strong><span>current feature dimensions</span></div>
+          <div class="stat"><strong>8,546</strong><span>feature dimensions</span></div>
           <div class="stat"><strong>12+12+4</strong><span>core, neural, and extension probes</span></div>
         </div>
       </div>
       <div class="hero-panel" aria-label="Signal summary">
         <div class="panel-top">
           <span>current feature allocation</span>
-          <span>window vector</span>
+          <span>aligned window</span>
         </div>
         <div class="signal"><code>mocap</code><div class="track"><span style="--w:24.8%;--c:#ccffa0"></span></div><strong>2,121</strong></div>
         <div class="signal"><code>camera+imu</code><div class="track"><span style="--w:1.5%;--c:#7ae5c3"></span></div><strong>126</strong></div>
@@ -2197,7 +2197,7 @@
             </article>
             <article class="brief-card">
               <strong>What comes next</strong>
-              <p>The next model-quality stage is not another single-sample score. It is a held-out episode pilot with at least 32 valid episodes, no train/test episode leakage, and a completed omni-model evaluation report.</p>
+              <p>The next model-quality stage is a held-out episode pilot over the selected multi-episode relay, with no train/test episode leakage and a completed omni-model evaluation report.</p>
             </article>
           </div>
           <div class="brief-actions">
@@ -2324,10 +2324,10 @@
           </article>
           <article class="roadmap-card" data-status="next">
             <span class="roadmap-status">next</span>
-            <h3>32-Episode Qwen3-Omni LoRA Pilot</h3>
-            <p>Train lightweight adapters and evaluate on held-out episodes with committed predictions, metrics, and run reports.</p>
+            <h3>Qwen3-Omni LoRA Pilot</h3>
+            <p>Train lightweight adapters on staged selected episodes and evaluate on held-out episodes with committed predictions, metrics, and run reports.</p>
             <div class="roadmap-meta">
-              <strong>Entry</strong><p>At least 32 valid staged episodes with no train/test episode leakage.</p>
+              <strong>Entry</strong><p>Selected episodes staged with no train/test episode leakage.</p>
               <strong>Evidence</strong><p>Dataset manifest, training metadata, progress logs, metrics, and predictions.</p>
             </div>
           </article>
@@ -2378,14 +2378,14 @@
           <p>The protocol is generated from committed metric artifacts so readers can see the exact data unit, split, task targets, leakage controls, and current limitations before comparing scores.</p>
         </div>
         <div class="artifact-grid">
-          <article class="artifact primary-artifact"><div><h3>Data unit</h3><p>One 20-frame aligned window from the public sample episode, stride 5 frames, 1,161 windows total, represented by the current 8,546-d feature vector.</p></div><a href="data/evaluation_protocol.json">protocol JSON</a></article>
+          <article class="artifact primary-artifact"><div><h3>Data unit</h3><p>One 20-frame aligned window from the public sample episode, stride 5 frames, 1,161 windows total, represented by 8,546 synchronized multimodal dimensions.</p></div><a href="data/evaluation_protocol.json">protocol JSON</a></article>
           <article class="artifact"><h3>Split policy</h3><p>Single-episode chronological 70/30 train/test split. This avoids random future-window mixing; cross-episode generalization is measured in the later multi-episode pilot.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/EVALUATION_PROTOCOL.md">protocol doc</a></article>
           <article class="artifact"><h3>Metric contract</h3><p>All 12 tasks list input, target, primary metric, minimal baseline score, and neural MLP score from committed result files.</p><a href="data/summary_metrics.json">summary metrics</a></article>
-          <article class="artifact"><h3>Leakage controls</h3><p>Scalers fit on train windows only; future labels, target feature blocks, caption/object labels, and contact labels stay on the target side unless explicitly queried.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/scripts/build_evaluation_protocol.py">builder script</a></article>
-          <article class="artifact"><h3>Audio ablation</h3><p>The current AAC block and a 588-d raw log-mel replacement are evaluated across all 12 task contracts under the same chronological split.</p><a href="data/audio_ablation_summary.json">audio summary</a></article>
+          <article class="artifact"><h3>Leakage controls</h3><p>Scalers fit on train windows only; future labels, target-side signals, caption/object labels, and contact labels stay on the target side unless explicitly queried.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/scripts/build_evaluation_protocol.py">builder script</a></article>
+          <article class="artifact"><h3>Audio ablation</h3><p>Audio and no-audio variants are evaluated across all 12 task contracts under the same chronological split.</p><a href="data/audio_ablation_summary.json">audio summary</a></article>
           <article class="artifact"><h3>Foundation branch selection</h3><p>Qwen3-Omni is the first trainable baseline, Cosmos 3 becomes the world-model branch, and policy models wait for explicit action targets.</p><a href="data/foundation_model_plan.json">backbone plan</a></article>
           <article class="artifact"><h3>Next evaluation stage</h3><p>This public-sample run covers single-episode task development. Cross-episode generalization, audio-visual learning, world modeling, policy targets, and held-out Qwen3-Omni training move to the multi-episode stage after selected data is staged.</p><a href="data/scope_claims_audit.json">pilot status</a></article>
-          <article class="artifact"><h3>Scale-up requirement</h3><p>The Omni pilot requires at least 32 valid episodes, held-out episode splits, no train/test episode leakage, training metadata, predictions, metrics, and a run report.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a></article>
+          <article class="artifact"><h3>Scale-up requirement</h3><p>The Omni pilot requires selected staged episodes, held-out episode splits, no train/test episode leakage, training metadata, predictions, metrics, and a run report.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a></article>
         </div>
       </div>
     </section>
@@ -2400,7 +2400,7 @@
           <article class="evidence-card">
             <span class="status-pill">verified</span>
             <h3>Aligned Xperience-10M sample windows</h3>
-            <p>5,821 frames become 1,161 synchronized 20-frame windows with an explicit 8,546-d feature contract.</p>
+            <p>5,821 frames become 1,161 synchronized 20-frame windows with an 8,546-dimensional representation.</p>
             <div class="evidence-links">
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/summary_report.json">summary_report.json</a>
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/feature_manifest.json">feature_manifest.json</a>
@@ -2418,7 +2418,7 @@
           <article class="evidence-card">
             <span class="status-pill">verified</span>
             <h3>Audio contribution is measured task by task</h3>
-            <p>Current AAC audio improves the primary metric on 6 of 12 task contracts; raw log-mel replacement improves over current audio on 6 of 12 tasks.</p>
+            <p>Audio variants improve the primary metric on 6 of 12 task contracts in this single-episode setting.</p>
             <div class="evidence-links">
               <a href="data/audio_ablation_summary.json">audio summary</a>
               <a href="assets/charts/audio_ablation_delta.svg">delta chart</a>
@@ -2455,7 +2455,7 @@
           <article class="evidence-card">
             <span class="status-pill">verified</span>
             <h3>Multi-episode pilot status is explicit</h3>
-            <p>The pilot status report records setup-stage <code>32ep</code> paths separately from completed held-out-episode metrics.</p>
+            <p>The pilot status report separates setup artifacts, selected relay state, and completed held-out-episode metrics.</p>
             <div class="evidence-links">
               <a href="data/scope_claims_audit.json">pilot status JSON</a>
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/scripts/validate_scope_claims.py">validator script</a>
@@ -2563,7 +2563,7 @@
           <article class="reading-card">
             <span class="step-index">02</span>
             <h3>Inspect one model input</h3>
-            <p>Use the window table and feature manifest to see the exact aligned sample unit, feature blocks, dimensions, and real AAC audio feature block.</p>
+            <p>Use the window table and feature manifest to see the aligned sample unit, modality sources, and leakage controls.</p>
             <div class="reading-links">
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/windows.csv">windows</a>
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/feature_manifest.json">features</a>
@@ -2590,7 +2590,7 @@
           </article>
         </div>
         <div class="boundary-strip">
-          <div class="boundary-item"><strong>Verified now</strong><span>One public episode, 5,821 frames, 1,161 windows, 8,546 current features, 12 minimal heads, 12 neural heads, and 4 direction-extension probes.</span></div>
+          <div class="boundary-item"><strong>Verified now</strong><span>One public episode, 5,821 frames, 1,161 aligned windows, 8,546 dimensions, 12 minimal heads, 12 neural heads, and 4 direction-extension probes.</span></div>
           <div class="boundary-item"><strong>Next: multi-episode</strong><span>A selected 128-episode held-out Qwen3-Omni LoRA pilot is being staged and must pass manifest, training, and evaluation checks before metrics are reported.</span></div>
           <div class="boundary-item"><strong>Not redistributed</strong><span>Raw videos, raw annotations, full Qwen weights, and private gated Xperience-10M data are not included in the public repo or HF bundles.</span></div>
         </div>
@@ -2611,7 +2611,7 @@
           <article class="artifact"><h3>Public sample card</h3><p>The sample repo lists <code>cc-by-nc-4.0</code>, HOMIE Toolkit for videos/annotations, and Rerun 0.29.0 for <code>.rrd</code> visualization.</p><a href="https://huggingface.co/datasets/ropedia-ai/xperience-10m-sample">sample dataset</a></article>
           <article class="artifact"><h3>Source notes</h3><p>The source notes summarize full-dataset facts, public sample-card facts, API-listing notes, and project coverage across the repo, website, and HF cards.</p><a href="data/source_alignment_audit.json">alignment report</a></article>
           <article class="artifact"><h3>Episode layout</h3><p>Expected folders contain six MP4 streams and <code>annotation.hdf5</code>; <code>visualization.rrd</code> is treated as a viewer artifact and excluded from training downloads.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/XPERIENCE10M_DATASET_CARD_ALIGNMENT.md">alignment note</a></article>
-          <article class="artifact"><h3>Current project subset</h3><p>One public sample episode, 5,821 frames, 1,161 windows, 8,546 current features including AAC audio, and no raw-data redistribution.</p><a href="data/modality_atlas.json">modality atlas</a></article>
+          <article class="artifact"><h3>Current project subset</h3><p>One public sample episode, 5,821 frames, 1,161 aligned windows, 8,546-dimensional task inputs, and no raw-data redistribution.</p><a href="data/modality_atlas.json">modality atlas</a></article>
           <article class="artifact"><h3>Covered now</h3><p>Action/subtask labels, next-action prediction, temporal diagnostics, hand trajectory, contact, object relevance, caption grounding, retrieval, reconstruction, and misalignment.</p><a href="data/summary_metrics.json">summary metrics</a></article>
           <article class="artifact"><h3>Responsible use</h3><p>The official card notes limited diversity and showcase/production quality. This project excludes identity, surveillance, biometric, sensitive-attribute, and safety-critical uses.</p><a href="data/xperience10m_dataset_card_alignment.json">use notes</a></article>
           <article class="artifact"><h3>Later milestones</h3><p>Full audio-visual learning, caption generation, depth-pixel prediction, SLAM estimation, neural rendering, policy learning, cross-episode generalization, and held-out Qwen3-Omni evaluation.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a></article>
@@ -2623,7 +2623,7 @@
       <div class="wrap">
         <div class="section-head">
           <h2>Ropedia Xperience-10M 12-task suite.</h2>
-          <p>The task map connects synchronized multimodal windows to 12 research task heads, then the modality atlas shows the sample streams used to build those contracts. AAC audio is decoded from the sample MP4 stream and included in the current 8,546-d baseline manifest.</p>
+          <p>The task map connects synchronized multimodal windows to 12 research task heads, then the modality atlas shows the sample streams used to build those contracts.</p>
         </div>
         <div class="figure-pan" id="task-suite-map">
           <img class="task-suite-image" src="assets/task_suite_infographic.png?v=xperience10m-taskfirst-v13-modality-xl" alt="Infographic showing all 12 Ropedia Xperience-10M tasks with enlarged full-width modality cards">
@@ -2645,12 +2645,12 @@
             <article class="atlas-card audio-card">
               <div class="atlas-top"><div><span class="atlas-index">02</span><h4>Audio</h4></div><span class="atlas-type">acoustic stream</span></div>
               <img src="assets/modalities/audio.png" alt="AAC waveform thumbnail from the public sample MP4 stream" loading="eager" decoding="async">
-              <div class="atlas-rows"><div class="atlas-row"><span>sample contains</span><p>AAC stream embedded in MP4</p></div><div class="atlas-row"><span>current baseline use</span><p>Decoded into a 168-d audio feature block</p></div></div>
+              <div class="atlas-rows"><div class="atlas-row"><span>sample contains</span><p>Audio stream embedded in MP4</p></div><div class="atlas-row"><span>current baseline use</span><p>Acoustic signal</p></div></div>
             </article>
             <article class="atlas-card">
               <div class="atlas-top"><div><span class="atlas-index">03</span><h4>Depth</h4></div><span class="atlas-type">geometry map</span></div>
               <img src="assets/modalities/depth.jpg" alt="Public sample depth and confidence thumbnails" loading="eager" decoding="async">
-              <div class="atlas-rows"><div class="atlas-row"><span>sample contains</span><p>Depth map + confidence channel</p></div><div class="atlas-row"><span>current baseline use</span><p>Spatial geometry feature block</p></div></div>
+              <div class="atlas-rows"><div class="atlas-row"><span>sample contains</span><p>Depth map + confidence channel</p></div><div class="atlas-row"><span>current baseline use</span><p>Spatial geometry signal</p></div></div>
             </article>
             <article class="atlas-card">
               <div class="atlas-top"><div><span class="atlas-index">04</span><h4>Pose / SLAM</h4></div><span class="atlas-type">camera pose</span></div>
@@ -2708,7 +2708,7 @@
           <article class="artifact primary-artifact">
             <div>
               <h3>One episode becomes a benchmark contract</h3>
-              <p>The public sample is converted into 5,821 frames, 1,161 aligned 20-frame windows, and an 8,546-dimensional feature contract.</p>
+              <p>The public sample is converted into 5,821 frames, 1,161 aligned 20-frame windows, and an 8,546-dimensional representation for repeatable task evaluation.</p>
             </div>
             <a href="data/research_takeaways.json">research_takeaways.json</a>
           </article>
@@ -2744,7 +2744,7 @@
         </div>
         <div class="models">
           <article class="model"><h3>Motion-only action</h3><span class="score">0.9688</span><span class="meta">macro-F1, 18 classes</span></article>
-          <article class="model"><h3>Current all-feature action</h3><span class="score">0.9829</span><span class="meta">macro-F1, 8,546 features</span></article>
+          <article class="model"><h3>Current all-feature action</h3><span class="score">0.9829</span><span class="meta">macro-F1, 8,546 dimensions</span></article>
           <article class="model"><h3>Motion-only subtask</h3><span class="score">0.9528</span><span class="meta">macro-F1, 14 classes</span></article>
           <article class="model"><h3>Current all-feature subtask</h3><span class="score">0.9173</span><span class="meta">macro-F1, chronological caveats</span></article>
         </div>
@@ -2756,7 +2756,7 @@
       <div class="wrap">
         <div class="section-head">
           <h2>Neural MLP heads, same task contracts.</h2>
-          <p>The neural baseline uses small PyTorch MLP classifiers/regressors on the same 8,546-d window features, chronological splits, and leakage filters. This isolates the value of a nonlinear head before moving to heavier Qwen/Omni experiments.</p>
+          <p>The neural baseline uses small PyTorch MLP classifiers/regressors on the same 8,546-dimensional windows, chronological splits, and leakage filters. This isolates the value of a nonlinear head before moving to heavier Qwen/Omni experiments.</p>
         </div>
         <div class="models">
           <article class="model"><h3>Neural hand forecast</h3><span class="score">0.1079</span><span class="meta">MPJPE, down from 0.8647 minimal</span></article>
@@ -2875,7 +2875,7 @@
       <div class="wrap">
         <div class="section-head">
           <h2>The 12 tasks share four head families.</h2>
-          <p>The diagram separates the shared episode-window feature pipeline from the task-specific heads. AAC audio is part of the current baseline feature block, and raw log-mel audio is now measured in the ablation upgrade.</p>
+          <p>The diagram separates the shared episode-window representation from the task-specific heads, so the task contracts stay readable before scaling to larger models.</p>
         </div>
         <img class="architecture-image" src="assets/task_architectures.png?v=xperience10m-nn" alt="Verified minimal and neural architecture diagram for all 12 Ropedia Xperience-10M tasks">
       </div>
@@ -2955,10 +2955,10 @@
     <section id="features" data-project-tab="method" role="tabpanel" aria-labelledby="tab-method" tabindex="-1">
       <div class="wrap">
         <div class="section-head">
-          <h2>Every feature block has a source.</h2>
-          <p>The point is not hidden complexity. Every block has a source modality, a dimensional footprint, and a manifest entry.</p>
+          <h2>Every model input has a source.</h2>
+          <p>The point is not hidden complexity. Every input group maps back to a source modality and a manifest entry.</p>
         </div>
-        <img class="chart" src="assets/charts/feature_blocks.svg" alt="All modality feature block chart">
+        <img class="chart" src="assets/charts/feature_blocks.svg" alt="All modality source chart">
       </div>
     </section>
 
@@ -2975,7 +2975,7 @@
           <img class="chart" src="assets/charts/episode_task_scores_minimal_vs_neural.svg" alt="Minimal versus neural score chart">
           <img class="chart" src="assets/charts/audio_ablation_delta.svg" alt="Measured audio delta chart across 12 task contracts">
         </div>
-        <p class="section-note"><a href="single_episode_explorer.html">Open the single-episode explorer</a> to inspect window-level labels, predictions, feature-block statistics, object labels, and diagnostic scores. The audio ablation report is available at <a href="data/audio_ablation_summary.json">audio_ablation_summary.json</a>.</p>
+        <p class="section-note"><a href="single_episode_explorer.html">Open the single-episode explorer</a> to inspect window-level labels, predictions, modality statistics, object labels, and diagnostic scores. The audio ablation report is available at <a href="data/audio_ablation_summary.json">audio_ablation_summary.json</a>.</p>
       </div>
     </section>
 
@@ -2989,7 +2989,7 @@
           <div class="content-tabs" role="tablist" aria-label="Artifact categories">
             <button type="button" class="content-tab active" id="artifact-tab-task-heads" role="tab" data-panel-target="artifact-panel-task-heads" aria-selected="true" aria-pressed="true" aria-controls="artifact-panel-task-heads">
               <strong>Task Heads</strong>
-              <span>windows, features, metrics</span>
+              <span>windows, tasks, metrics</span>
             </button>
             <button type="button" class="content-tab" id="artifact-tab-public-surfaces" role="tab" data-panel-target="artifact-panel-public-surfaces" aria-selected="false" aria-pressed="false" aria-controls="artifact-panel-public-surfaces" tabindex="-1">
               <strong>Public Surfaces</strong>
@@ -3007,18 +3007,18 @@
           <section class="artifact-group tabbed-panel" id="artifact-panel-task-heads" role="tabpanel" aria-labelledby="artifact-tab-task-heads">
             <div class="artifact-group-head">
               <div><span>Research artifacts</span><h3>From one episode to task heads</h3></div>
-              <p>Start with the files that define the sample windows, feature blocks, task contracts, metrics, walkthroughs, and research-direction mapping.</p>
+              <p>Start with the files that define the sample windows, modality inputs, task contracts, metrics, walkthroughs, and research-direction mapping.</p>
             </div>
             <div class="artifact-grid">
               <article class="artifact primary-artifact"><div><h3>Task-suite report</h3><p>One JSON file with every task definition, split detail, feature dimension, and minimal/neural metric.</p></div><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/summary_report.json">summary_report.json</a></article>
               <article class="artifact"><h3>Windows table</h3><p>Window start/end frames and aligned action/subtask labels for the public sample episode.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/windows.csv">windows.csv</a></article>
-              <article class="artifact"><h3>Feature manifest</h3><p>Start/end index and dimension for every current feature block in the 8,546-d window vector.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/feature_manifest.json">feature_manifest.json</a></article>
+              <article class="artifact"><h3>Feature manifest</h3><p>Technical source map for the current modality inputs used by the task suite.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/feature_manifest.json">feature_manifest.json</a></article>
               <article class="artifact"><h3>Neural MLP task results</h3><p>Per-task PyTorch MLP metrics, predictions, histories, and checkpoints for the same 12 task contracts.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/episode_task_suite/neural_mlp">neural_mlp/</a></article>
               <article class="artifact"><h3>Four-direction taxonomy</h3><p>Generated JSON, CSV, Markdown, and website data mapping all 12 tasks to the four research tracks.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/episode_task_suite/research_directions">research_directions/</a></article>
               <article class="artifact"><h3>Direction extension probes</h3><p>Four coded probes, one per research direction, with minimal and neural metrics plus prediction/rank CSVs.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/episode_task_suite/research_direction_extensions">research_direction_extensions/</a></article>
               <article class="artifact"><h3>Task walkthroughs</h3><p>Case studies for all 12 tasks, including input, middle process modules, output, metric, limitation, and task-player data.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/episode_task_suite/task_walkthroughs">task_walkthroughs/</a></article>
-              <article class="artifact"><h3>Audio ablation and raw upgrade</h3><p>All 72 task/variant rows comparing current audio, no audio, handcrafted-audio only, raw-audio only, raw replacement, and all-plus-raw.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/audio_ablation">audio_ablation/</a></article>
-              <article class="artifact"><h3>Single-episode explorer</h3><p>Interactive window-level view of labels, predictions, feature-block statistics, object labels, and diagnostics.</p><a href="single_episode_explorer.html">single_episode_explorer.html</a></article>
+              <article class="artifact"><h3>Audio ablation and raw upgrade</h3><p>All 72 task/variant rows comparing current audio, no audio, raw audio, replacement, and combined-input settings.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/audio_ablation">audio_ablation/</a></article>
+              <article class="artifact"><h3>Single-episode explorer</h3><p>Interactive window-level view of labels, predictions, modality statistics, object labels, and diagnostics.</p><a href="single_episode_explorer.html">single_episode_explorer.html</a></article>
               <article class="artifact"><h3>Cross-modal retrieval</h3><p>The strongest self-supervised signal from the single episode.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/cross_modal_retrieval/metrics.json">metrics.json</a></article>
             </div>
           </section>
@@ -3067,7 +3067,7 @@
               <article class="artifact"><h3>Release checks</h3><p>One release map for automated validators and live post-publish checks.</p><a href="data/quality_gates.json">quality_gates.json</a></article>
               <article class="artifact"><h3>Mirror parity</h3><p>Prepared repo, HF Space, artifact dataset, and model bundle parity for critical data, figures, website HTML, and validator files.</p><a href="data/mirror_parity.json">mirror_parity.json</a></article>
               <article class="artifact"><h3>Live publication</h3><p>Last public GitHub/HF URL verification after upload.</p><a href="data/live_publication_status.json">live_publication_status.json</a></article>
-              <article class="artifact"><h3>Multi-episode pilot status</h3><p>Records setup-stage <code>32ep</code> identifiers separately from completed held-out-episode results.</p><a href="data/scope_claims_audit.json">scope_claims_audit.json</a></article>
+              <article class="artifact"><h3>Multi-episode pilot status</h3><p>Separates setup artifacts, selected relay state, and completed held-out-episode results.</p><a href="data/scope_claims_audit.json">scope_claims_audit.json</a></article>
               <article class="artifact"><h3>Public project surface</h3><p>Presents repo, website, and Hugging Face cards with consistent naming, links, tab semantics, and reader-facing copy.</p><a href="data/public_surface_qa.json">public_surface_qa.json</a></article>
               <article class="artifact"><h3>Public bundle contents</h3><p>Summarizes raw-data exclusion, cache exclusion, archive exclusion, token-string checks, and public figure references.</p><a href="data/publication_audit.json">publication_audit.json</a></article>
             </div>

index.html

@@ -2121,14 +2121,14 @@
         <div class="hero-stats">
           <div class="stat"><strong>5,821</strong><span>frames in sample episode</span></div>
           <div class="stat"><strong>1,161</strong><span>20-frame windows</span></div>
-          <div class="stat"><strong>8,546</strong><span>current feature dimensions</span></div>
+          <div class="stat"><strong>8,546</strong><span>feature dimensions</span></div>
           <div class="stat"><strong>12+12+4</strong><span>core, neural, and extension probes</span></div>
         </div>
       </div>
       <div class="hero-panel" aria-label="Signal summary">
         <div class="panel-top">
           <span>current feature allocation</span>
-          <span>window vector</span>
+          <span>aligned window</span>
         </div>
         <div class="signal"><code>mocap</code><div class="track"><span style="--w:24.8%;--c:#ccffa0"></span></div><strong>2,121</strong></div>
         <div class="signal"><code>camera+imu</code><div class="track"><span style="--w:1.5%;--c:#7ae5c3"></span></div><strong>126</strong></div>
@@ -2197,7 +2197,7 @@
             </article>
             <article class="brief-card">
               <strong>What comes next</strong>
-              <p>The next model-quality stage is not another single-sample score. It is a held-out episode pilot with at least 32 valid episodes, no train/test episode leakage, and a completed omni-model evaluation report.</p>
+              <p>The next model-quality stage is a held-out episode pilot over the selected multi-episode relay, with no train/test episode leakage and a completed omni-model evaluation report.</p>
             </article>
           </div>
           <div class="brief-actions">
@@ -2265,12 +2265,12 @@
             </div>
           </article>
           <article class="snapshot-card gated">
-            <span class="status-pill">data-gated</span>
+            <span class="status-pill">staging</span>
             <h3>Omni-model scale-up path</h3>
-            <p>The 32-episode LoRA path is prepared; full training results require gated data access, held-out splits, training, and evaluation.</p>
+            <p>Full-dataset access is granted, a 128-episode relay is in progress, and full training results require completed staging, held-out splits, training, and evaluation.</p>
             <div class="snapshot-meta">
-              <span>current stage <strong>setup checked</strong></span>
-              <span>target gate <strong>32 episodes</strong></span>
+              <span>current stage <strong>relay started</strong></span>
+              <span>selected set <strong>128 episodes</strong></span>
               <span>held-out eval <strong>pending</strong></span>
             </div>
           </article>
@@ -2324,10 +2324,10 @@
           </article>
           <article class="roadmap-card" data-status="next">
             <span class="roadmap-status">next</span>
-            <h3>32-Episode Qwen3-Omni LoRA Pilot</h3>
-            <p>Train lightweight adapters and evaluate on held-out episodes with committed predictions, metrics, and run reports.</p>
+            <h3>Qwen3-Omni LoRA Pilot</h3>
+            <p>Train lightweight adapters on staged selected episodes and evaluate on held-out episodes with committed predictions, metrics, and run reports.</p>
             <div class="roadmap-meta">
-              <strong>Entry</strong><p>At least 32 valid staged episodes with no train/test episode leakage.</p>
+              <strong>Entry</strong><p>Selected episodes staged with no train/test episode leakage.</p>
               <strong>Evidence</strong><p>Dataset manifest, training metadata, progress logs, metrics, and predictions.</p>
             </div>
           </article>
@@ -2336,7 +2336,7 @@
             <h3>Foundation-Model Selection Matrix</h3>
             <p>Keep Qwen3-Omni as the first trainable held-out pilot, add Cosmos 3 for world modeling, and stage policy candidates after action targets are explicit.</p>
             <div class="roadmap-meta">
-              <strong>Entry</strong><p>32-episode data gate or a 3-8 episode preprocessing dry run.</p>
+              <strong>Entry</strong><p>Completed 128-episode staging or a smaller 3-8 episode preprocessing dry run.</p>
               <strong>Evidence</strong><p>Foundation model plan, source links, model-specific entry conditions, and evaluation additions.</p>
             </div>
           </article>
@@ -2345,7 +2345,7 @@
             <h3>64-128 Episode Robustness Run</h3>
             <p>Test whether pilot conclusions survive broader sessions, missing modalities, and stronger ablations.</p>
             <div class="roadmap-meta">
-              <strong>Entry</strong><p>32-episode pilot trains and evaluates cleanly.</p>
+              <strong>Entry</strong><p>Selected multi-episode pilot trains and evaluates cleanly.</p>
               <strong>Evidence</strong><p>Metrics by session, task, modality, ablation, and failure type.</p>
             </div>
           </article>
@@ -2378,14 +2378,14 @@
           <p>The protocol is generated from committed metric artifacts so readers can see the exact data unit, split, task targets, leakage controls, and current limitations before comparing scores.</p>
         </div>
         <div class="artifact-grid">
-          <article class="artifact primary-artifact"><div><h3>Data unit</h3><p>One 20-frame aligned window from the public sample episode, stride 5 frames, 1,161 windows total, represented by the current 8,546-d feature vector.</p></div><a href="data/evaluation_protocol.json">protocol JSON</a></article>
+          <article class="artifact primary-artifact"><div><h3>Data unit</h3><p>One 20-frame aligned window from the public sample episode, stride 5 frames, 1,161 windows total, represented by 8,546 synchronized multimodal dimensions.</p></div><a href="data/evaluation_protocol.json">protocol JSON</a></article>
           <article class="artifact"><h3>Split policy</h3><p>Single-episode chronological 70/30 train/test split. This avoids random future-window mixing; cross-episode generalization is measured in the later multi-episode pilot.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/EVALUATION_PROTOCOL.md">protocol doc</a></article>
           <article class="artifact"><h3>Metric contract</h3><p>All 12 tasks list input, target, primary metric, minimal baseline score, and neural MLP score from committed result files.</p><a href="data/summary_metrics.json">summary metrics</a></article>
-          <article class="artifact"><h3>Leakage controls</h3><p>Scalers fit on train windows only; future labels, target feature blocks, caption/object labels, and contact labels stay on the target side unless explicitly queried.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/scripts/build_evaluation_protocol.py">builder script</a></article>
-          <article class="artifact"><h3>Audio ablation</h3><p>The current AAC block and a 588-d raw log-mel replacement are evaluated across all 12 task contracts under the same chronological split.</p><a href="data/audio_ablation_summary.json">audio summary</a></article>
+          <article class="artifact"><h3>Leakage controls</h3><p>Scalers fit on train windows only; future labels, target-side signals, caption/object labels, and contact labels stay on the target side unless explicitly queried.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/scripts/build_evaluation_protocol.py">builder script</a></article>
+          <article class="artifact"><h3>Audio ablation</h3><p>Audio and no-audio variants are evaluated across all 12 task contracts under the same chronological split.</p><a href="data/audio_ablation_summary.json">audio summary</a></article>
           <article class="artifact"><h3>Foundation branch selection</h3><p>Qwen3-Omni is the first trainable baseline, Cosmos 3 becomes the world-model branch, and policy models wait for explicit action targets.</p><a href="data/foundation_model_plan.json">backbone plan</a></article>
-          <article class="artifact"><h3>Next evaluation stage</h3><p>This public-sample run covers single-episode task development. Cross-episode generalization, audio-visual learning, world modeling, policy targets, and full 32-episode Qwen3-Omni training move to the multi-episode stage.</p><a href="data/scope_claims_audit.json">pilot status</a></article>
-          <article class="artifact"><h3>Scale-up requirement</h3><p>The Omni pilot requires at least 32 valid episodes, held-out episode splits, no train/test episode leakage, training metadata, predictions, metrics, and a run report.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a></article>
+          <article class="artifact"><h3>Next evaluation stage</h3><p>This public-sample run covers single-episode task development. Cross-episode generalization, audio-visual learning, world modeling, policy targets, and held-out Qwen3-Omni training move to the multi-episode stage after selected data is staged.</p><a href="data/scope_claims_audit.json">pilot status</a></article>
+          <article class="artifact"><h3>Scale-up requirement</h3><p>The Omni pilot requires selected staged episodes, held-out episode splits, no train/test episode leakage, training metadata, predictions, metrics, and a run report.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a></article>
         </div>
       </div>
     </section>
@@ -2400,7 +2400,7 @@
           <article class="evidence-card">
             <span class="status-pill">verified</span>
             <h3>Aligned Xperience-10M sample windows</h3>
-            <p>5,821 frames become 1,161 synchronized 20-frame windows with an explicit 8,546-d feature contract.</p>
+            <p>5,821 frames become 1,161 synchronized 20-frame windows with an 8,546-dimensional representation.</p>
             <div class="evidence-links">
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/summary_report.json">summary_report.json</a>
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/feature_manifest.json">feature_manifest.json</a>
@@ -2418,7 +2418,7 @@
           <article class="evidence-card">
             <span class="status-pill">verified</span>
             <h3>Audio contribution is measured task by task</h3>
-            <p>Current AAC audio improves the primary metric on 6 of 12 task contracts; raw log-mel replacement improves over current audio on 6 of 12 tasks.</p>
+            <p>Audio variants improve the primary metric on 6 of 12 task contracts in this single-episode setting.</p>
             <div class="evidence-links">
               <a href="data/audio_ablation_summary.json">audio summary</a>
               <a href="assets/charts/audio_ablation_delta.svg">delta chart</a>
@@ -2444,9 +2444,9 @@
             </div>
           </article>
           <article class="evidence-card">
-            <span class="status-pill">data-gated</span>
+            <span class="status-pill">staging</span>
             <h3>Qwen3-Omni pilot setup</h3>
-            <p>The current Qwen3-Omni artifacts use one episode and 128 train windows. The 32-episode evaluation is still pending.</p>
+            <p>The current Qwen3-Omni artifacts use one episode and 128 train windows. A 128-episode selected relay is in progress for held-out evaluation.</p>
             <div class="evidence-links">
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/EVIDENCE_CONTRACT.md">evidence contract</a>
               <a href="data/evidence_contract.json">machine JSON</a>
@@ -2455,7 +2455,7 @@
           <article class="evidence-card">
             <span class="status-pill">verified</span>
             <h3>Multi-episode pilot status is explicit</h3>
-            <p>The pilot status report records setup-stage <code>32ep</code> paths separately from completed held-out-episode metrics.</p>
+            <p>The pilot status report separates setup artifacts, selected relay state, and completed held-out-episode metrics.</p>
             <div class="evidence-links">
               <a href="data/scope_claims_audit.json">pilot status JSON</a>
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/scripts/validate_scope_claims.py">validator script</a>
@@ -2563,7 +2563,7 @@
           <article class="reading-card">
             <span class="step-index">02</span>
             <h3>Inspect one model input</h3>
-            <p>Use the window table and feature manifest to see the exact aligned sample unit, feature blocks, dimensions, and real AAC audio feature block.</p>
+            <p>Use the window table and feature manifest to see the aligned sample unit, modality sources, and leakage controls.</p>
             <div class="reading-links">
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/windows.csv">windows</a>
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/feature_manifest.json">features</a>
@@ -2581,7 +2581,7 @@
           <article class="reading-card">
             <span class="step-index">04</span>
             <h3>Check the scale-up gate</h3>
-            <p>The multi-episode Qwen3-Omni path is prepared. The 32-episode result will be added after the data gate and held-out evaluation pass.</p>
+            <p>The multi-episode Qwen3-Omni path is prepared. The selected 128-episode result will be added after staging, preprocessing, training, and held-out evaluation pass.</p>
             <div class="reading-links">
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a>
               <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md">access status</a>
@@ -2590,8 +2590,8 @@
           </article>
         </div>
         <div class="boundary-strip">
-          <div class="boundary-item"><strong>Verified now</strong><span>One public episode, 5,821 frames, 1,161 windows, 8,546 current features, 12 minimal heads, 12 neural heads, and 4 direction-extension probes.</span></div>
-          <div class="boundary-item"><strong>Next: multi-episode</strong><span>A 32-episode held-out Qwen3-Omni LoRA pilot is gated on Xperience-10M access and must pass manifest, training, and evaluation checks.</span></div>
+          <div class="boundary-item"><strong>Verified now</strong><span>One public episode, 5,821 frames, 1,161 aligned windows, 8,546 dimensions, 12 minimal heads, 12 neural heads, and 4 direction-extension probes.</span></div>
+          <div class="boundary-item"><strong>Next: multi-episode</strong><span>A selected 128-episode held-out Qwen3-Omni LoRA pilot is being staged and must pass manifest, training, and evaluation checks before metrics are reported.</span></div>
           <div class="boundary-item"><strong>Not redistributed</strong><span>Raw videos, raw annotations, full Qwen weights, and private gated Xperience-10M data are not included in the public repo or HF bundles.</span></div>
         </div>
       </div>
@@ -2611,10 +2611,10 @@
           <article class="artifact"><h3>Public sample card</h3><p>The sample repo lists <code>cc-by-nc-4.0</code>, HOMIE Toolkit for videos/annotations, and Rerun 0.29.0 for <code>.rrd</code> visualization.</p><a href="https://huggingface.co/datasets/ropedia-ai/xperience-10m-sample">sample dataset</a></article>
           <article class="artifact"><h3>Source notes</h3><p>The source notes summarize full-dataset facts, public sample-card facts, API-listing notes, and project coverage across the repo, website, and HF cards.</p><a href="data/source_alignment_audit.json">alignment report</a></article>
           <article class="artifact"><h3>Episode layout</h3><p>Expected folders contain six MP4 streams and <code>annotation.hdf5</code>; <code>visualization.rrd</code> is treated as a viewer artifact and excluded from training downloads.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/XPERIENCE10M_DATASET_CARD_ALIGNMENT.md">alignment note</a></article>
-          <article class="artifact"><h3>Current project subset</h3><p>One public sample episode, 5,821 frames, 1,161 windows, 8,546 current features including AAC audio, and no raw-data redistribution.</p><a href="data/modality_atlas.json">modality atlas</a></article>
+          <article class="artifact"><h3>Current project subset</h3><p>One public sample episode, 5,821 frames, 1,161 aligned windows, 8,546-dimensional task inputs, and no raw-data redistribution.</p><a href="data/modality_atlas.json">modality atlas</a></article>
           <article class="artifact"><h3>Covered now</h3><p>Action/subtask labels, next-action prediction, temporal diagnostics, hand trajectory, contact, object relevance, caption grounding, retrieval, reconstruction, and misalignment.</p><a href="data/summary_metrics.json">summary metrics</a></article>
           <article class="artifact"><h3>Responsible use</h3><p>The official card notes limited diversity and showcase/production quality. This project excludes identity, surveillance, biometric, sensitive-attribute, and safety-critical uses.</p><a href="data/xperience10m_dataset_card_alignment.json">use notes</a></article>
-          <article class="artifact"><h3>Later milestones</h3><p>Full audio-visual learning, caption generation, depth-pixel prediction, SLAM estimation, neural rendering, policy learning, cross-episode generalization, and 32-episode Qwen3-Omni evaluation.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a></article>
+          <article class="artifact"><h3>Later milestones</h3><p>Full audio-visual learning, caption generation, depth-pixel prediction, SLAM estimation, neural rendering, policy learning, cross-episode generalization, and held-out Qwen3-Omni evaluation.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">data status</a></article>
         </div>
       </div>
     </section>
@@ -2623,7 +2623,7 @@
       <div class="wrap">
         <div class="section-head">
           <h2>Ropedia Xperience-10M 12-task suite.</h2>
-          <p>The task map connects synchronized multimodal windows to 12 research task heads, then the modality atlas shows the sample streams used to build those contracts. AAC audio is decoded from the sample MP4 stream and included in the current 8,546-d baseline manifest.</p>
+          <p>The task map connects synchronized multimodal windows to 12 research task heads, then the modality atlas shows the sample streams used to build those contracts.</p>
         </div>
         <div class="figure-pan" id="task-suite-map">
           <img class="task-suite-image" src="assets/task_suite_infographic.png?v=xperience10m-taskfirst-v13-modality-xl" alt="Infographic showing all 12 Ropedia Xperience-10M tasks with enlarged full-width modality cards">
@@ -2645,12 +2645,12 @@
             <article class="atlas-card audio-card">
               <div class="atlas-top"><div><span class="atlas-index">02</span><h4>Audio</h4></div><span class="atlas-type">acoustic stream</span></div>
               <img src="assets/modalities/audio.png" alt="AAC waveform thumbnail from the public sample MP4 stream" loading="eager" decoding="async">
-              <div class="atlas-rows"><div class="atlas-row"><span>sample contains</span><p>AAC stream embedded in MP4</p></div><div class="atlas-row"><span>current baseline use</span><p>Decoded into a 168-d audio feature block</p></div></div>
+              <div class="atlas-rows"><div class="atlas-row"><span>sample contains</span><p>Audio stream embedded in MP4</p></div><div class="atlas-row"><span>current baseline use</span><p>Acoustic signal</p></div></div>
             </article>
             <article class="atlas-card">
               <div class="atlas-top"><div><span class="atlas-index">03</span><h4>Depth</h4></div><span class="atlas-type">geometry map</span></div>
               <img src="assets/modalities/depth.jpg" alt="Public sample depth and confidence thumbnails" loading="eager" decoding="async">
-              <div class="atlas-rows"><div class="atlas-row"><span>sample contains</span><p>Depth map + confidence channel</p></div><div class="atlas-row"><span>current baseline use</span><p>Spatial geometry feature block</p></div></div>
+              <div class="atlas-rows"><div class="atlas-row"><span>sample contains</span><p>Depth map + confidence channel</p></div><div class="atlas-row"><span>current baseline use</span><p>Spatial geometry signal</p></div></div>
             </article>
             <article class="atlas-card">
               <div class="atlas-top"><div><span class="atlas-index">04</span><h4>Pose / SLAM</h4></div><span class="atlas-type">camera pose</span></div>
@@ -2708,7 +2708,7 @@
           <article class="artifact primary-artifact">
             <div>
               <h3>One episode becomes a benchmark contract</h3>
-              <p>The public sample is converted into 5,821 frames, 1,161 aligned 20-frame windows, and an 8,546-dimensional feature contract.</p>
+              <p>The public sample is converted into 5,821 frames, 1,161 aligned 20-frame windows, and an 8,546-dimensional representation for repeatable task evaluation.</p>
             </div>
             <a href="data/research_takeaways.json">research_takeaways.json</a>
           </article>
@@ -2729,7 +2729,7 @@
           </article>
           <article class="artifact">
             <h3>Scale means held-out episodes</h3>
-            <p>The next credible model-quality unit is a 32-episode held-out pilot across 32 sessions, not more adjacent windows from one sample.</p>
+            <p>The next credible model-quality unit is a held-out multi-episode pilot across different sessions, not more adjacent windows from one sample.</p>
             <a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md">scale-up status</a>
           </article>
         </div>
@@ -2744,7 +2744,7 @@
         </div>
         <div class="models">
           <article class="model"><h3>Motion-only action</h3><span class="score">0.9688</span><span class="meta">macro-F1, 18 classes</span></article>
-          <article class="model"><h3>Current all-feature action</h3><span class="score">0.9829</span><span class="meta">macro-F1, 8,546 features</span></article>
+          <article class="model"><h3>Current all-feature action</h3><span class="score">0.9829</span><span class="meta">macro-F1, 8,546 dimensions</span></article>
           <article class="model"><h3>Motion-only subtask</h3><span class="score">0.9528</span><span class="meta">macro-F1, 14 classes</span></article>
           <article class="model"><h3>Current all-feature subtask</h3><span class="score">0.9173</span><span class="meta">macro-F1, chronological caveats</span></article>
         </div>
@@ -2756,7 +2756,7 @@
       <div class="wrap">
         <div class="section-head">
           <h2>Neural MLP heads, same task contracts.</h2>
-          <p>The neural baseline uses small PyTorch MLP classifiers/regressors on the same 8,546-d window features, chronological splits, and leakage filters. This isolates the value of a nonlinear head before moving to heavier Qwen/Omni experiments.</p>
+          <p>The neural baseline uses small PyTorch MLP classifiers/regressors on the same 8,546-dimensional windows, chronological splits, and leakage filters. This isolates the value of a nonlinear head before moving to heavier Qwen/Omni experiments.</p>
         </div>
         <div class="models">
           <article class="model"><h3>Neural hand forecast</h3><span class="score">0.1079</span><span class="meta">MPJPE, down from 0.8647 minimal</span></article>
@@ -2875,7 +2875,7 @@
       <div class="wrap">
         <div class="section-head">
           <h2>The 12 tasks share four head families.</h2>
-          <p>The diagram separates the shared episode-window feature pipeline from the task-specific heads. AAC audio is part of the current baseline feature block, and raw log-mel audio is now measured in the ablation upgrade.</p>
+          <p>The diagram separates the shared episode-window representation from the task-specific heads, so the task contracts stay readable before scaling to larger models.</p>
         </div>
         <img class="architecture-image" src="assets/task_architectures.png?v=xperience10m-nn" alt="Verified minimal and neural architecture diagram for all 12 Ropedia Xperience-10M tasks">
       </div>
@@ -2955,10 +2955,10 @@
     <section id="features" data-project-tab="method" role="tabpanel" aria-labelledby="tab-method" tabindex="-1">
       <div class="wrap">
         <div class="section-head">
-          <h2>Every feature block has a source.</h2>
-          <p>The point is not hidden complexity. Every block has a source modality, a dimensional footprint, and a manifest entry.</p>
+          <h2>Every model input has a source.</h2>
+          <p>The point is not hidden complexity. Every input group maps back to a source modality and a manifest entry.</p>
         </div>
-        <img class="chart" src="assets/charts/feature_blocks.svg" alt="All modality feature block chart">
+        <img class="chart" src="assets/charts/feature_blocks.svg" alt="All modality source chart">
       </div>
     </section>
 
@@ -2975,7 +2975,7 @@
           <img class="chart" src="assets/charts/episode_task_scores_minimal_vs_neural.svg" alt="Minimal versus neural score chart">
           <img class="chart" src="assets/charts/audio_ablation_delta.svg" alt="Measured audio delta chart across 12 task contracts">
         </div>
-        <p class="section-note"><a href="single_episode_explorer.html">Open the single-episode explorer</a> to inspect window-level labels, predictions, feature-block statistics, object labels, and diagnostic scores. The audio ablation report is available at <a href="data/audio_ablation_summary.json">audio_ablation_summary.json</a>.</p>
+        <p class="section-note"><a href="single_episode_explorer.html">Open the single-episode explorer</a> to inspect window-level labels, predictions, modality statistics, object labels, and diagnostic scores. The audio ablation report is available at <a href="data/audio_ablation_summary.json">audio_ablation_summary.json</a>.</p>
       </div>
     </section>
 
@@ -2989,7 +2989,7 @@
           <div class="content-tabs" role="tablist" aria-label="Artifact categories">
             <button type="button" class="content-tab active" id="artifact-tab-task-heads" role="tab" data-panel-target="artifact-panel-task-heads" aria-selected="true" aria-pressed="true" aria-controls="artifact-panel-task-heads">
               <strong>Task Heads</strong>
-              <span>windows, features, metrics</span>
+              <span>windows, tasks, metrics</span>
             </button>
             <button type="button" class="content-tab" id="artifact-tab-public-surfaces" role="tab" data-panel-target="artifact-panel-public-surfaces" aria-selected="false" aria-pressed="false" aria-controls="artifact-panel-public-surfaces" tabindex="-1">
               <strong>Public Surfaces</strong>
@@ -2997,7 +2997,7 @@
             </button>
             <button type="button" class="content-tab" id="artifact-tab-scale-up" role="tab" data-panel-target="artifact-panel-scale-up" aria-selected="false" aria-pressed="false" aria-controls="artifact-panel-scale-up" tabindex="-1">
               <strong>Scale-Up</strong>
-              <span>data gate and Omni path</span>
+              <span>relay and Omni path</span>
             </button>
             <button type="button" class="content-tab" id="artifact-tab-checks" role="tab" data-panel-target="artifact-panel-checks" aria-selected="false" aria-pressed="false" aria-controls="artifact-panel-checks" tabindex="-1">
               <strong>Checks</strong>
@@ -3007,18 +3007,18 @@
           <section class="artifact-group tabbed-panel" id="artifact-panel-task-heads" role="tabpanel" aria-labelledby="artifact-tab-task-heads">
             <div class="artifact-group-head">
               <div><span>Research artifacts</span><h3>From one episode to task heads</h3></div>
-              <p>Start with the files that define the sample windows, feature blocks, task contracts, metrics, walkthroughs, and research-direction mapping.</p>
+              <p>Start with the files that define the sample windows, modality inputs, task contracts, metrics, walkthroughs, and research-direction mapping.</p>
             </div>
             <div class="artifact-grid">
               <article class="artifact primary-artifact"><div><h3>Task-suite report</h3><p>One JSON file with every task definition, split detail, feature dimension, and minimal/neural metric.</p></div><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/summary_report.json">summary_report.json</a></article>
               <article class="artifact"><h3>Windows table</h3><p>Window start/end frames and aligned action/subtask labels for the public sample episode.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/windows.csv">windows.csv</a></article>
-              <article class="artifact"><h3>Feature manifest</h3><p>Start/end index and dimension for every current feature block in the 8,546-d window vector.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/feature_manifest.json">feature_manifest.json</a></article>
+              <article class="artifact"><h3>Feature manifest</h3><p>Technical source map for the current modality inputs used by the task suite.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/feature_manifest.json">feature_manifest.json</a></article>
               <article class="artifact"><h3>Neural MLP task results</h3><p>Per-task PyTorch MLP metrics, predictions, histories, and checkpoints for the same 12 task contracts.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/episode_task_suite/neural_mlp">neural_mlp/</a></article>
               <article class="artifact"><h3>Four-direction taxonomy</h3><p>Generated JSON, CSV, Markdown, and website data mapping all 12 tasks to the four research tracks.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/episode_task_suite/research_directions">research_directions/</a></article>
               <article class="artifact"><h3>Direction extension probes</h3><p>Four coded probes, one per research direction, with minimal and neural metrics plus prediction/rank CSVs.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/episode_task_suite/research_direction_extensions">research_direction_extensions/</a></article>
               <article class="artifact"><h3>Task walkthroughs</h3><p>Case studies for all 12 tasks, including input, middle process modules, output, metric, limitation, and task-player data.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/episode_task_suite/task_walkthroughs">task_walkthroughs/</a></article>
-              <article class="artifact"><h3>Audio ablation and raw upgrade</h3><p>All 72 task/variant rows comparing current audio, no audio, handcrafted-audio only, raw-audio only, raw replacement, and all-plus-raw.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/audio_ablation">audio_ablation/</a></article>
-              <article class="artifact"><h3>Single-episode explorer</h3><p>Interactive window-level view of labels, predictions, feature-block statistics, object labels, and diagnostics.</p><a href="single_episode_explorer.html">single_episode_explorer.html</a></article>
+              <article class="artifact"><h3>Audio ablation and raw upgrade</h3><p>All 72 task/variant rows comparing current audio, no audio, raw audio, replacement, and combined-input settings.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/tree/main/results/audio_ablation">audio_ablation/</a></article>
+              <article class="artifact"><h3>Single-episode explorer</h3><p>Interactive window-level view of labels, predictions, modality statistics, object labels, and diagnostics.</p><a href="single_episode_explorer.html">single_episode_explorer.html</a></article>
               <article class="artifact"><h3>Cross-modal retrieval</h3><p>The strongest self-supervised signal from the single episode.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/episode_task_suite/cross_modal_retrieval/metrics.json">metrics.json</a></article>
             </div>
           </section>
@@ -3043,14 +3043,14 @@
           <section class="artifact-group tabbed-panel" id="artifact-panel-scale-up" role="tabpanel" aria-labelledby="artifact-tab-scale-up" hidden>
             <div class="artifact-group-head">
               <div><span>Scale-up path</span><h3>Prepared for multi-episode training</h3></div>
-              <p>The multi-episode Qwen3-Omni path is documented and scripted. Full-pilot metrics come after the data gate and held-out evaluation pass.</p>
+              <p>The multi-episode Qwen3-Omni path is documented and scripted. Full-pilot metrics come after selected data is staged and held-out evaluation passes.</p>
             </div>
             <div class="artifact-grid">
-              <article class="artifact primary-artifact"><div><h3>Project scope</h3><p>Connects implemented single-episode artifacts, setup-stage Omni work, pending data access, and later multi-episode milestones.</p></div><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/EVIDENCE_CONTRACT.md">EVIDENCE_CONTRACT.md</a></article>
+              <article class="artifact primary-artifact"><div><h3>Project scope</h3><p>Connects implemented single-episode artifacts, setup-stage Omni work, current 128-episode relay, and later multi-episode milestones.</p></div><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/EVIDENCE_CONTRACT.md">EVIDENCE_CONTRACT.md</a></article>
               <article class="artifact"><h3>Foundation-model plan</h3><p>Backbone selection matrix covering Qwen3-Omni, Cosmos 3, GR00T, OpenVLA/openpi, Gemini Robotics, Octo, and SmolVLA-style policy candidates.</p><a href="data/foundation_model_plan.json">foundation_model_plan.json</a></article>
-              <article class="artifact"><h3>Multi-episode access status</h3><p>Public data-access path, selected 32-episode pilot plan, and data requirements.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md">MULTI_EPISODE_ACCESS_STATUS.md</a></article>
+              <article class="artifact"><h3>Multi-episode access status</h3><p>Public data-access path, selected 128-episode relay plan, and data requirements.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md">MULTI_EPISODE_ACCESS_STATUS.md</a></article>
               <article class="artifact"><h3>Qwen3-Omni setup artifacts</h3><p>Manifests, metadata, metrics, and progress logs from the current setup run.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/episode_manifest.json">episode_manifest.json</a></article>
-              <article class="artifact"><h3>32-episode data requirement</h3><p>The data status file defines what must be available before full pilot training and held-out metrics.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">DATA_ACCESS_STATUS.md</a></article>
+              <article class="artifact"><h3>Multi-episode data requirement</h3><p>The data status file defines what must be available before full pilot training and held-out metrics.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">DATA_ACCESS_STATUS.md</a></article>
             </div>
           </section>
 
@@ -3067,7 +3067,7 @@
               <article class="artifact"><h3>Release checks</h3><p>One release map for automated validators and live post-publish checks.</p><a href="data/quality_gates.json">quality_gates.json</a></article>
               <article class="artifact"><h3>Mirror parity</h3><p>Prepared repo, HF Space, artifact dataset, and model bundle parity for critical data, figures, website HTML, and validator files.</p><a href="data/mirror_parity.json">mirror_parity.json</a></article>
               <article class="artifact"><h3>Live publication</h3><p>Last public GitHub/HF URL verification after upload.</p><a href="data/live_publication_status.json">live_publication_status.json</a></article>
-              <article class="artifact"><h3>Multi-episode pilot status</h3><p>Records setup-stage <code>32ep</code> identifiers separately from completed held-out-episode results.</p><a href="data/scope_claims_audit.json">scope_claims_audit.json</a></article>
+              <article class="artifact"><h3>Multi-episode pilot status</h3><p>Separates setup artifacts, selected relay state, and completed held-out-episode results.</p><a href="data/scope_claims_audit.json">scope_claims_audit.json</a></article>
               <article class="artifact"><h3>Public project surface</h3><p>Presents repo, website, and Hugging Face cards with consistent naming, links, tab semantics, and reader-facing copy.</p><a href="data/public_surface_qa.json">public_surface_qa.json</a></article>
               <article class="artifact"><h3>Public bundle contents</h3><p>Summarizes raw-data exclusion, cache exclusion, archive exclusion, token-string checks, and public figure references.</p><a href="data/publication_audit.json">publication_audit.json</a></article>
             </div>
@@ -3079,14 +3079,14 @@
     <section id="omni-relay" data-project-tab="resources" role="tabpanel" aria-labelledby="tab-resources" tabindex="-1">
       <div class="wrap">
         <div class="section-head">
-          <h2>Qwen3-Omni pilot is approval-ready.</h2>
-          <p>The full Xperience-10M Hugging Face dataset is gated. While access is pending, the public plan has selected a 32-episode pilot across 32 different session UUIDs.</p>
+          <h2>Qwen3-Omni pilot is in data staging.</h2>
+          <p>Full Xperience-10M access is granted. The current plan selects 128 metadata-balanced episodes across 128 different session UUIDs, with raw staging in progress and no held-out metrics reported yet.</p>
         </div>
         <div class="artifact-grid">
-          <article class="artifact"><h3>Selection</h3><p>Stratified round-robin over 64 top-level sessions; 680 complete candidates scanned; 32 sessions selected.</p></article>
+          <article class="artifact"><h3>Selection</h3><p>128 complete episodes selected from 128 unique top-level sessions, balanced across episode-size bands and split 96/16/16 for train/val/test.</p></article>
           <article class="artifact"><h3>Transfer</h3><p>Download raw episodes only from official gated sources, exclude visualization.rrd, validate files, then stage them for training.</p></article>
-          <article class="artifact"><h3>Current LoRA artifact</h3><p>The current LoRA artifact uses the locally available sample data. The 32-episode result begins after gated data is staged and held-out evaluation runs.</p></article>
-          <article class="artifact"><h3>Backbone branches</h3><p>Qwen3-Omni is the immediate LoRA path; Cosmos 3 is the first world-model branch; GR00T/OpenVLA/openpi become policy branches after action targets are auditable.</p><a href="data/foundation_model_plan.json">backbone plan</a></article>
+          <article class="artifact"><h3>Current LoRA artifact</h3><p>The current LoRA artifact uses the locally available sample data. The multi-episode result begins after selected data is staged, preprocessed, trained, and evaluated on held-out sessions.</p></article>
+          <article class="artifact"><h3>Backbone branches</h3><p>Qwen3-Omni is the immediate LoRA path; Cosmos 3 is the first world-model branch; GR00T/OpenVLA/openpi become policy branches after action targets are well-defined.</p><a href="data/foundation_model_plan.json">backbone plan</a></article>
         </div>
       </div>
     </section>
@@ -3102,7 +3102,7 @@
           <article class="artifact"><h3>Reproducibility matrix</h3><p>Machine-readable command matrix covering sample download, baselines, 12 tasks, figures, and validation.</p><a href="data/reproducibility_matrix.json">reproducibility_matrix.json</a></article>
           <article class="artifact"><h3>Exact-match reproduction record</h3><p>The last metric rebuild reproduced the public-sample outputs from a fresh cache and matched the committed metrics.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/notes/reproducibility_audit.md">reproducibility_audit.md</a></article>
           <article class="artifact"><h3>Website reference report</h3><p>Local HTML references, anchors, JSON bundles, and image dimensions are validated before publishing.</p><a href="data/website_integrity.json">website_integrity.json</a></article>
-          <article class="artifact"><h3>32-Episode pilot status</h3><p>The 32-episode Qwen3-Omni pilot is prepared at the code and selection-plan level; final metrics follow gated data access and held-out evaluation.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">DATA_ACCESS_STATUS.md</a></article>
+          <article class="artifact"><h3>Multi-episode pilot status</h3><p>The Qwen3-Omni pilot is prepared at the code and selection-plan level; final metrics follow completed staging, preprocessing, training, and held-out evaluation.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/results/omni_finetune/DATA_ACCESS_STATUS.md">DATA_ACCESS_STATUS.md</a></article>
         </div>
         <p class="repro-note">Minimal path: install the toolkit dependencies, download the official sample, run the 12-task suite with neural heads, regenerate visualizations, then run the artifact index and publication validator.</p>
         <pre class="code-panel"><button type="button" data-copy="setup">Copy</button><code id="setup">git clone https://github.com/Ropedia/HOMIE-toolkit.git

metrics/artifact_index.json

@@ -1,6 +1,6 @@
 {
   "title": "Ropedia Xperience-10M Task Suite Artifact Index",
-  "generated_at_utc": "2026-06-03T17:18:29+00:00",
+  "generated_at_utc": "2026-06-03T18:11:35+00:00",
   "status": "pass",
   "artifact_count": 72,
   "missing": [],
@@ -40,8 +40,8 @@
       "surface": "repo_hf",
       "shows": "Gives first-pass readers a concise project shape before the detailed artifact trail.",
       "exists": true,
-      "bytes": 2494,
-      "sha256": "54c6906103f6d7f45398cd638730c3eb9813d6ca4a807920d6edba4b12f446e6"
+      "bytes": 2516,
+      "sha256": "ace07e60bab3d17d17d2af32b9843582adfbb8d0abe5f7c0489c1e9502eab2f8"
     },
     {
       "id": "project_brief_json",
@@ -51,8 +51,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable first-reader project brief for the website and Hugging Face mirrors.",
       "exists": true,
-      "bytes": 2563,
-      "sha256": "c8df06022de6a0d51be993175ac63764a607f48c944f1d9451c5bbeebf6cef54"
+      "bytes": 2585,
+      "sha256": "f0e2e7853b86cdcd2f06269d75f19a96432a782efdf99c08aca7d689fd2cd0a1"
     },
     {
       "id": "project_status",
@@ -62,8 +62,8 @@
       "surface": "repo_hf",
       "shows": "Gives a compact current-state table for first-pass readers.",
       "exists": true,
-      "bytes": 6721,
-      "sha256": "166795e9f5986015f7a9e74934fbd878f5dcc7cbac79d49483bd514feacbfde8"
+      "bytes": 6648,
+      "sha256": "c873c0361ee1b83a835f1112529a557e2033959d70a9bfe726a66cdd120a9047"
     },
     {
       "id": "project_status_json",
@@ -73,8 +73,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable copy of the current project status for website and HF mirrors.",
       "exists": true,
-      "bytes": 9006,
-      "sha256": "38d289ce14e7b4f0998ce156718662f2298ee3aa4e77782a40316ad33aefaefb"
+      "bytes": 8868,
+      "sha256": "ea2deebfe7dd2ee7cb141623f4c6c1d4d287bb94a1c642609b854e934d5d701a"
     },
     {
       "id": "research_roadmap",
@@ -84,8 +84,8 @@
       "surface": "repo_hf",
       "shows": "Defines the staged path from public-sample task development to multi-episode held-out evaluation and larger omni-model extensions.",
       "exists": true,
-      "bytes": 6677,
-      "sha256": "590702b8000bd143fdbcf65104b0cad835e6e4c79754c28c31727f1bdf43ae79"
+      "bytes": 6645,
+      "sha256": "b8bebfded058cc2e3f9c918d99600c878f015da08861801fb3ebae73b85fdc57"
     },
     {
       "id": "research_roadmap_json",
@@ -95,8 +95,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable staged roadmap for the website and Hugging Face mirrors.",
       "exists": true,
-      "bytes": 5752,
-      "sha256": "0510e919ad9ffa3aebe052273e90133a57572c3b21b991db1ff28de83061f6ca"
+      "bytes": 5726,
+      "sha256": "903ec0d6104c092fa0356c7741134fee4a1c78f3b0517001c49ce11f8077d6e2"
     },
     {
       "id": "foundation_model_plan",
@@ -106,8 +106,8 @@
       "surface": "repo_hf",
       "shows": "Defines the post-data-gate backbone choices: Qwen3-Omni first, Cosmos 3 for world modeling, and VLA/policy models after action-target conversion.",
       "exists": true,
-      "bytes": 6538,
-      "sha256": "15431b3ed368d660e777655d2e4a97f2870997f2ad38b451a73d631b5e431bcf"
+      "bytes": 6545,
+      "sha256": "f7d35f38db371d52b5bb92e59b951b93946c425fe12bf283f69f869a45a5c60f"
     },
     {
       "id": "foundation_model_plan_json",
@@ -117,8 +117,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable foundation-model selection matrix with source links, entry conditions, and evaluation additions.",
       "exists": true,
-      "bytes": 8883,
-      "sha256": "0465df50071ab4c8e46acafdd2fa1653f23dacd33a1e1e1d65a571c82348ddbc"
+      "bytes": 8880,
+      "sha256": "a11476cc697e7bc2c29b3db569d059c8f536025fdd0b47e49f060a88664fd450"
     },
     {
       "id": "evidence_contract",
@@ -128,8 +128,8 @@
       "surface": "repo",
       "shows": "Defines the implemented scope, setup-stage items, and multi-episode prerequisites.",
       "exists": true,
-      "bytes": 11172,
-      "sha256": "7197fe7e42ea4cbe200b203871d0db745c201c37eef088243aa21ac4a72fe782"
+      "bytes": 11287,
+      "sha256": "46b03a5c31ec31476c0476b11bc72375d786a5e8ea11fcba81e1500a0e50b9b8"
     },
     {
       "id": "project_packet",
@@ -139,8 +139,8 @@
       "surface": "website_hf",
       "shows": "Gives a short project path with scope status and public surfaces.",
       "exists": true,
-      "bytes": 7659,
-      "sha256": "7d9f5ecc8adf9b3913e99a8aec28372529e63c105739fa53b3f8a7a9470470c2"
+      "bytes": 7553,
+      "sha256": "2f2d227bf3d95eeba7e4f5133d37fbe62910cd373aa3431d9b960914a590e149"
     },
     {
       "id": "artifact_guide",
@@ -150,8 +150,8 @@
       "surface": "repo_hf",
       "shows": "Gives the human-readable map from project scope to data, tasks, platform mirrors, and scale-up status.",
       "exists": true,
-      "bytes": 16974,
-      "sha256": "e456d291614db14a2cb6a91cabf1b044f47c518d39d7d701eb191f9434f2c157"
+      "bytes": 16876,
+      "sha256": "6b392ff36290f8eefab393218c17aff4d90ba28457324089f78e7a4882b17dbc"
     },
     {
       "id": "official_dataset_card_alignment",
@@ -161,8 +161,8 @@
       "surface": "repo_hf",
       "shows": "Aligns public dataset wording with the official gated Xperience-10M card, public sample card, HF API metadata, and current project coverage.",
       "exists": true,
-      "bytes": 10764,
-      "sha256": "8eaada6926194aeccf5af6742098f9be8aa611fc9bdf73cfa2df806327b89e9c"
+      "bytes": 10625,
+      "sha256": "9e51442cb13cb4074d73fb212b49c56e6e4ae6dea10fd135a4020f7488e49be0"
     },
     {
       "id": "official_dataset_card_alignment_json",
@@ -172,8 +172,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable upstream dataset-card, sample-card, and HF API alignment facts for website and HF mirrors.",
       "exists": true,
-      "bytes": 7585,
-      "sha256": "8bb8f22cf1cc8938f7dcfef15d11e3e544243bb56dc32c792313d224e91a8a2d"
+      "bytes": 7559,
+      "sha256": "24ea3fc9e1ae8f698ca192ae1b8511dfe1129b807af237f41382e0644e5b58af"
     },
     {
       "id": "source_alignment",
@@ -238,8 +238,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable protocol generated from committed task metrics for website and HF mirrors.",
       "exists": true,
-      "bytes": 13644,
-      "sha256": "69281b1727d3de7a734fe89846e9881b02fa1df4108f4ae39946acbcdf6ed8bf"
+      "bytes": 13625,
+      "sha256": "33a9589ad613bf524c5db932660b810cb724c8765d8bfa65a9996b01286e39fa"
     },
     {
       "id": "evaluation_protocol_builder",
@@ -249,8 +249,8 @@
       "surface": "repo_hf",
       "shows": "Regenerates the protocol from committed summary metrics and task artifacts.",
       "exists": true,
-      "bytes": 16163,
-      "sha256": "b352798b50212dac792855313bdc0ad1e5e2ef835de1ff25b03d52afd1d33f03"
+      "bytes": 16139,
+      "sha256": "e2aaac38f451402971e03c004d32d5c5f8a05d67dc1820d6bf9b12d1da1a292f"
     },
     {
       "id": "research_takeaways",
@@ -260,8 +260,8 @@
       "surface": "repo_hf",
       "shows": "Summarizes the main research lessons from committed metrics and identifies which experiments need held-out episodes.",
       "exists": true,
-      "bytes": 4893,
-      "sha256": "16a002167d0702ea7e85ad1a494800daad4ff080c3751e078275f7c196721046"
+      "bytes": 4889,
+      "sha256": "1f688343561501898fdc615f8ac67ac1ae64fc2acc4290b3a17a5b4f0c732b99"
     },
     {
       "id": "research_takeaways_json",
@@ -271,8 +271,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable result interpretation for the website, HF cards, and mirror checks.",
       "exists": true,
-      "bytes": 6814,
-      "sha256": "538ec80950dbf5d77cc6ab2bd5a96a3af191b56a257b895e48b6cf28deb0d044"
+      "bytes": 6821,
+      "sha256": "509d8690a4854097ea45af635dafa0231efa11838f1bb1f9cb338e99009a0206"
     },
     {
       "id": "research_takeaways_builder",
@@ -282,19 +282,19 @@
       "surface": "repo_hf",
       "shows": "Regenerates the research takeaways from committed summary metrics and task result artifacts.",
       "exists": true,
-      "bytes": 13304,
-      "sha256": "135f555b17aae5dc5965ec88c3db6e49622a480467909ffa10d53278d4ea054b"
+      "bytes": 13285,
+      "sha256": "19baa1dbe3e4454e040bb1b2fc711e1dd5ce5280ee19bfc01b01cca854006d38"
     },
     {
       "id": "audio_ablation_script",
-      "title": "Audio ablation and raw-audio upgrade script",
+      "title": "Audio contribution script",
       "path": "scripts/audio_ablation_and_raw_upgrade.py",
       "kind": "result_interpretation",
       "surface": "repo_hf",
-      "shows": "Measures current AAC audio contribution and a raw log-mel audio feature replacement across all 12 task contracts.",
+      "shows": "Measures audio contribution variants across all 12 task contracts.",
       "exists": true,
-      "bytes": 43192,
-      "sha256": "73db45a3ddb50b60f61ca5e55bdf2008e48c268681bfcdc95d35bb56e48f89e3"
+      "bytes": 43144,
+      "sha256": "f7e3a38ec906dac7ca55b13c49720bd41ed89a1fd994c7d54730a4de5dfd1b59"
     },
     {
       "id": "audio_ablation_summary",
@@ -302,7 +302,7 @@
       "path": "results/audio_ablation/audio_ablation_summary.json",
       "kind": "metrics_source",
       "surface": "repo_hf",
-      "shows": "Stores per-task audio deltas for all current features, no-audio, handcrafted-audio-only, raw-audio-only, raw replacement, and all-plus-raw variants.",
+      "shows": "Stores per-task audio deltas for all current features, no-audio, audio-only, alternate-audio-only, replacement, and all-plus-alternate variants.",
       "exists": true,
       "bytes": 9735,
       "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
@@ -313,7 +313,7 @@
       "path": "results/audio_ablation/AUDIO_ABLATION_SUMMARY.md",
       "kind": "result_interpretation",
       "surface": "repo_hf",
-      "shows": "Human-readable table showing the measured audio contribution and raw-audio replacement delta for every task.",
+      "shows": "Human-readable table showing the measured audio contribution and alternate-representation delta for every task.",
       "exists": true,
       "bytes": 2166,
       "sha256": "9318bc55b1b051c384ce2fd6e7cd0a56becfa00185a0b597fdc474a40b1f17f9"
@@ -326,8 +326,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable audio ablation summary mirrored into the static website and Hugging Face bundles.",
       "exists": true,
-      "bytes": 9735,
-      "sha256": "514370c53baeb51373160bbd5ab7f5f0de7733301bf1b82981ed799aec41152a"
+      "bytes": 9701,
+      "sha256": "7debd9e67b7df09322d743193bf8b785277a992c9e8c82c08c7a36c4e066e6de"
     },
     {
       "id": "audio_ablation_delta_chart",
@@ -414,8 +414,8 @@
       "surface": "repo_hf",
       "shows": "Lists the automated and post-publish checks used to keep the release current.",
       "exists": true,
-      "bytes": 4919,
-      "sha256": "7138e99b116c44f128cd2f749e9d7427e496cb7596d30d54dd44af79be80df81"
+      "bytes": 4901,
+      "sha256": "7426ef148491d7288161a8d13cf31e81d5351ab1c920614a59b06888889c97d3"
     },
     {
       "id": "quality_gate_manifest",
@@ -425,8 +425,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable release-check summary for validators, mirrors, and public project surfaces.",
       "exists": true,
-      "bytes": 8147,
-      "sha256": "067b2b8e1ed1be2e869e11c1203a4069293dae13445f4844ee59a511de6a3d50"
+      "bytes": 8118,
+      "sha256": "9c40b314c53d8406e52288656e699ad707a7cc2aa5161bdb5117f1a5b84e4ea2"
     },
     {
       "id": "public_surface_qa",
@@ -551,8 +551,8 @@
       "surface": "repo_hf",
       "shows": "Defines public reproduction commands, expected outputs, and non-reproducible scale-up boundaries.",
       "exists": true,
-      "bytes": 6312,
-      "sha256": "99a2df715ce481f49991de620f20e5093f37767fa36122c1505c74ee6ea92e3c"
+      "bytes": 6346,
+      "sha256": "48d82bbdfeba88a63100f55e19ef7e1392c0e633b5f3f0c648444f9ed1c96e02"
     },
     {
       "id": "reproducibility_matrix",
@@ -562,8 +562,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable reproduction steps with expected artifacts and public boundaries.",
       "exists": true,
-      "bytes": 5197,
-      "sha256": "c3d353084ee641ff3d1369dd911577ff7a1b5d7b2faf841faa12caa69dfa5ed0"
+      "bytes": 5229,
+      "sha256": "928eb95b95d4028a7f21135584c37ff141833e26178cc918a874a6448f8c5ded"
     },
     {
       "id": "artifact_index_builder",
@@ -573,8 +573,8 @@
       "surface": "repo_hf",
       "shows": "Generates the selective artifact catalog from local files.",
       "exists": true,
-      "bytes": 26652,
-      "sha256": "faf1e0b56eb9ba4bc94e04bb1aeb9a392fb4410e8053564dd43bce3d8aacfa4d"
+      "bytes": 26566,
+      "sha256": "982fcc36e6dfda3382662d1f1e1f539f27f45aaddd6503aad77a14145bcc1bf6"
     },
     {
       "id": "publication_audit",
@@ -595,7 +595,7 @@
       "kind": "scale_up_status",
       "surface": "website_hf",
       "volatile": true,
-      "shows": "Records historical 32ep setup paths separately from completed held-out-episode results.",
+      "shows": "Separates setup paths from completed held-out-episode results.",
       "exists": true,
       "bytes": 20066,
       "hash_policy": "existence_and_size_only"
@@ -654,8 +654,8 @@
       "surface": "website_hf",
       "shows": "Mirrors task metrics for the static dashboard.",
       "exists": true,
-      "bytes": 25210,
-      "sha256": "a88302490b9099e7d148a8a6dd0ba6393d7ef26a0ba2c6a0edf552da94b948a8"
+      "bytes": 25198,
+      "sha256": "a5c337bc6be321a9999dd69f0920756ad722d505697827425a64005f05a74699"
     },
     {
       "id": "feature_manifest",
@@ -709,8 +709,8 @@
       "surface": "repo_hf",
       "shows": "Maps the 12 tasks to the four Ropedia research directions as direct/proxy/diagnostic.",
       "exists": true,
-      "bytes": 14429,
-      "sha256": "292ddd47b5f50cc5a6a0d4099466b134774a1ca7610afe50a2a5781c6c181bf9"
+      "bytes": 14414,
+      "sha256": "7215681c55a6739da7f16d833f62fcb8f6d58069840963182e98427eaf0cf654"
     },
     {
       "id": "research_direction_extensions",
@@ -720,8 +720,8 @@
       "surface": "repo_hf",
       "shows": "Stores one coded extension probe per research direction with minimal and neural metrics.",
       "exists": true,
-      "bytes": 11907,
-      "sha256": "aa65022a8a41af4f0bca5197add932c4f3f51b13b8324323280f64457473ff03"
+      "bytes": 11904,
+      "sha256": "b72bb62f95d27de296ae526568c55bca3aeb9afe687e57bbe652ca74a32c93d7"
     },
     {
       "id": "task_walkthroughs",
@@ -731,8 +731,8 @@
       "surface": "repo_hf",
       "shows": "Explains every task with case study, input, process modules, output, and limitation.",
       "exists": true,
-      "bytes": 15449,
-      "sha256": "9309a4f6da85c43c9d87be6ed0458c3cbc9c01b664da6610cbbb2392d282302e"
+      "bytes": 15436,
+      "sha256": "de8607242ca8c5c1c632f7175eeffd0c009536eb050614bfae2ff561e14cd92d"
     },
     {
       "id": "task_suite_infographic",
@@ -753,8 +753,8 @@
       "surface": "website_hf",
       "shows": "Documents the seven public-sample modality cards and their derived thumbnail assets.",
       "exists": true,
-      "bytes": 3819,
-      "sha256": "b1285148588ef402b70c830b62d66fbbcfa86bc5ffc2ad99e165a1429bb49084"
+      "bytes": 3797,
+      "sha256": "4f12a5471d4fe53b1f5c1ef6f9535cc14974690fa6ffbfbeddc8e1b90f86b281"
     },
     {
       "id": "modality_thumbnails",
@@ -795,10 +795,10 @@
       "path": "results/omni_finetune/DATA_ACCESS_STATUS.md",
       "kind": "scaleup_status",
       "surface": "repo_hf",
-      "shows": "Summarizes the data-access requirement before the 32-episode Qwen3-Omni pilot can run.",
+      "shows": "Summarizes the staging requirement before the held-out Qwen3-Omni pilot can report metrics.",
       "exists": true,
-      "bytes": 2954,
-      "sha256": "6ac14781581f06004bdc2fb0e2425c616419b738e6b858f8611bd19fab80741c"
+      "bytes": 2940,
+      "sha256": "c33c40b2bdda0a057cbfef28e1fc75695e4e7273703fc261137f796fded27d47"
     },
     {
       "id": "multi_episode_access_status",
@@ -808,8 +808,8 @@
       "surface": "repo_hf",
       "shows": "Documents the public multi-episode access status and 32-episode pilot selection.",
       "exists": true,
-      "bytes": 2282,
-      "sha256": "e270cdfa4485458572a40f5c3c4bd598fe53e2b8de60af3ff1364398d96862e2"
+      "bytes": 2285,
+      "sha256": "e05b3396bb7c1254ab516d154f7deca293be188d3a48624862822b2b5e26dc91"
     },
     {
       "id": "citation",