UNITAS: A 3D-Native World Action Model for Embodied Manipulation
Paper • 2610.12099 • Published • 6
Finn Reyes
finnreyes
AI & ML interests
combinatorial optimization, test-time compute
Recent Activity
upvoted a paper 1 day ago
UNITAS: A 3D-Native World Action Model for Embodied Manipulation upvoted a paper 1 day ago
SpatialASM: Internalizing Active Mentalization for Spatial ReasoningOrganizations
None yet
upvoted 5
papers 1 day ago
Relevance Does Not Imply Applicability: Experience Activation for Personal GUI Agents
Paper • 2609.33304 • Published • 45
SpatialASM: Internalizing Active Mentalization for Spatial Reasoning
Paper • 2610.13832 • Published • 29
T3lescope: Arbitrary-Resolution High-Fidelity Generative Surface Reconstruction from Images
Paper • 2610.03308 • Published • 16
SPLATIFY: Reproduce, Discover, Innovate! From Papers and Ideas to Trainable 3DGS Code
Paper • 2610.09116 • Published • 10
upvoted 7
papers 2 days ago
Multi-Stream Attention Residuals
Paper • 2610.13630 • Published • 47
VIS-Ground: Video Interactive Storytelling with Contextual Grounding
Paper • 2610.09326 • Published • 99
GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed
Paper • 2609.39600 • Published • 98
CrossTimeEdit: A Decade-Spanning Cross-View Dataset and Reward-Guided Editing for Historical Street-View Generation
Paper • 2609.36616 • Published • 33
Visual Grounding Safety in Vision-Language Models
Paper • 2610.05637 • Published • 11
GroundingPI: A Grounding Foundation Model towards Physical Intelligence with Visual Primitives
Paper • 2609.39601 • Published • 320
PR-OPD: Privileged Representation On-policy Self-Distillation for Agentic Reinforcement Learning
Paper • 2609.36642 • Published • 42
upvoted 5
papers 3 days ago
Post-Training Frontier Text-to-Image Models by Composing Preference and Rubric Rewards
Paper • 2610.02967 • Published • 42
LEGO: A Lifting-Free Approach for Exocentric-to-Egocentric Video Generation
Paper • 2610.12442 • Published • 41
One Block, Multiple Depths: Recurrent Vision Transformers with Depth-Programmed Experts
Paper • 2610.12448 • Published • 11
REMORY: Learning Residual Memory for Context Compaction
Paper • 2610.11287 • Published • 61
SparseEngine: Sparse-First Inference Engine
Paper • 2609.39068 • Published • 22
upvoted 3
papers 4 days ago