tardellirs (Tardelli Stekel)

Posts 3

Post

3

Paper Pulse: the history Hugging Face doesn't show for Daily Papers 📈

On Hugging Face, a paper shows one number: its upvotes today. You can't see how it got there, where it ranked on its day, or whether anyone actually uses it.

Paper Pulse keeps that history for all 18,897 Daily Papers since March 2024: every paper's upvote curve, day by day, its rank on its day, week and month, and the models on the Hub that cite it.

With the history, things show up that the Hub hides 👀
⚖️ Upvotes ≠ use. "Soaring from 4K to 400K" got 30 upvotes, and the models citing it were downloaded 87.7M times last month. YaRN: 88 upvotes, 80.2M. Meanwhile, 6 of the 8 most upvoted papers of the past year (470+ upvotes each) aren't cited by a single model yet.
🏁 The first day decides: on 82% of days, the paper that finished #1 was already #1 when its first day ended.

What's in it:
📊 every paper's upvote curve, its rank on its day, week and month, and the models citing it with their downloads
⏱️ the hour-by-hour race for today's top spot
🔥 Most used papers: Daily Papers ranked by real use
🎁 Wrapped for any author or lab: your last 12 months on Daily Papers, as a card to share. NVIDIA tops 1,389 labs, with 5K+ upvotes and 6 papers of the day. Get yours: tardellirs/paper-pulse
🏷️ a live badge for your README

🔗 App: tardellirs/paper-pulse
📦 Open data: tardellirs/paper-pulse-data

Built on @hysts 's daily-papers-stats snapshots. Thanks for making them public! 🙏 And thanks @akhaliq for Daily Papers itself.

Download numbers come from its sister project, Model Pulse, which tracks every model on the Hub: tardellirs/model-pulse

Post

3823

Robotics is now the second most downloaded dataset category on the Hub, after text generation.

Robotics datasets got 13.7M downloads in September, ahead of text classification and question answering. Two years ago the category ranked 23rd. One in 7 new datasets is now robotics, mostly LeRobot recordings: typically a few dozen demos, about half of them on low-cost SO-100/SO-101 arms.

I found this after adding datasets and Spaces to Model Pulse, which rebuilds daily history from @cfahlgren1 's hub-stats snapshots. Two more findings:

- In 2022, 36% of authors who list training data cited classic NLP sets like IMDb, SQuAD and GLUE. In 2026 it's 2.4%. Reasoning traces distilled from models like DeepSeek-R1 and Claude are now the most cited kind.
- In October 2025, 122K Spaces were created, 71K of them websites built with DeepSite. That's about 6x the monthly pace of late 2024, while likes given per month fell from about 35K to about 20K.

New in the app: a page for every dataset, with daily downloads and the models trained on it (636 list FineWeb), a page for every Space, and rankings for both.

Thank you to everyone who liked Model Pulse this week: it made Spaces of the Week and is #7 on trending. Thanks also to @dipankarsarkar , whose comments on the last post fixed three data issues. If a number looks wrong, tell me.

Spaces: tardellirs/model-pulse