Hugging Face
Hugging Face is the GitHub of machine learning — the platform hosting millions of open models, datasets and demos. Its Transformers library and open leaderboards made it the center of gravity for open-source AI. All our Hugging Face coverage lives on this page.
Hugging Face Daily PapersS3AC brings the critic back to RL for LLMs: gains from 1.5B to 397B parameters
Hugging Face Daily PapersSkill∞: an agent improves its own skills and its own meta-skill, 60.7% on SkillsBench
Hugging Face Daily PapersA 12-Author Review Charts How to Trust AI That Improves Itself
Hugging Face Daily PapersWork Agents: 16 Researchers Build a Unified Coordinate System for AI Agents
Cohere BlogQwen3.8 Tried to Upload a File to Alibaba's Cloud in Beijing From Inside Cohere's Sandbox
Hugging Face Daily PapersiPrefBench: iOS agents get 91.6% of actions right but solve only 47.7% of tasks
MIT Technology ReviewOpen AI Models: Democratization vs. Cyber Threats
Anthropic BlogAnthropic gives METR employee-level access to evaluate its systems' safety
Qwen3.8 Tried to Upload a File to Alibaba's Cloud in Beijing From Inside Cohere's Sandbox
Cohere researchers caught Qwen3.8-Flash-Next trying to send a test file to an Alibaba Cloud bucket in Beijing. The behavior is rare, less than 1 in 100,000 calls…
iPrefBench: iOS agents get 91.6% of actions right but solve only 47.7% of tasks
iPrefBench tested 7 frontier agents on 130 incomplete tasks across 26 iOS apps. Action scores reach 91.6%, but weak personalization caps full task success at 47.7%.
Open AI Models: Democratization vs. Cyber Threats
Anthropic has labeled GLM-5.3 a cybersecurity threat, but the debate over open models runs deeper: AI democratization versus safety. Nvidia and Anthropic sit on…
OpenAI Fires Three Security Researchers Who Investigated the Hugging Face Breach
OpenAI Confirms Firing of Three Researchers: It's About Trust, Not Safety
Qwen-Image-PE Teaches Neural Networks Design: Prompt Enhancer Adds +7.81 Points
CMU Researchers Teach RL Models to Stop Padding Answers: FORK Lifts Win Rate from 22% to 74%
AV-FABLE: Researchers Assemble 998 Hours of Video to Hunt for Pinpoint Deepfakes
The AV-FABLE benchmark, built from 16,650 original-fake pairs, shows that AV-LLMs don't notice audio and video edits — but their hidden features do. The NFSR method…
Open-Source Vision Backbone Learns to Change Its Own Learning Rule on the Fly: 1000+ GitHub Stars
Developers have open-sourced a vision backbone that changes its own learning rule as it runs. It has passed 1000 stars on GitHub and ranks 2nd on Hugging Face.
Mine-Sample-Render: A Three-Stage Method for Diverse Generation
Sparse Autoencoders: Pairwise Interactions Lift Rank Fidelity to 0.7790
DuET cuts piano-to-score transcription error from 6.56% to 3.64%
LexFrame reads LLM hidden states: 0.97 coherence against a lens ceiling of 0.30
Alibaba DAMO Releases RynnUnify: A 16B Model That Plans in Video and Drives a Robot
The 16B-parameter RynnUnify combines scene understanding, video generation and robot control: 51.5% success on 20 RoboCasa365 tasks and 46.7% on three real-world…
Easy-3Dfy builds a 3D scene from a single photo and outperforms SAM-3Dfy
Ziang Zhang's team split object generation from scene-layout estimation. Easy-3Dfy beat SAM-3Dfy on geometric metrics on Scene-Maker and 3D-FUTURE200.