Research Papers
EvoSkill-GUI Lets GUI Agents Rewrite Their Own Skills at Deployment, No Training Required
A new Hugging Face paper proposes EvoSkill-GUI, a training-free framework that treats GUI agent skills as living, multi-file packa...
Hugging Face Paper Unveils Fuse: A Verifiable Testbed for LLM Social Reasoning
Researchers introduce Fuse, a multi-agent simulation framework that creates verifiable ground truth for evaluating how LLM assista...
Hugging Face Paper Traces On-Policy Distillation Length Inflation to EOS Token Mismatch
A new Hugging Face paper finds that base students and post-trained teachers can favor different end-of-sequence tokens, causing st...
RetireOPD Lets RL Agents Fire Their Teacher When It Stops Helping
A new Hugging Face paper proposes RetireOPD, a self-retiring on-policy distillation method in which skill-free agents decide on th...
RiskChainBench Tests Whether AI Can Decode Hidden Abuse Messages and Investigate the Sites They Point To
A new Hugging Face benchmark pairs 3,600 obfuscated message-restoration inputs with 600 human-labeled local web environments, scor...
DeepSeek-V4.1-Flash Cuts KV Cache to 890 Bytes per Token
DeepSeek's new multimodal Mixture-of-Experts model pairs a Causal Encoder-Decoder design with CSA2 cross-layer reuse and FP4 KV ca...
JEPA-Anything: One Predictive Framework for Seven Scientific Worlds
Hugging Face researchers introduce JEPA-Anything, a domain-agnostic world-modeling framework built on orthogonal predictive factor...
Hugging Face Paper Unveils SoL-Pi, a Token-Efficient Agent Harness
A new Hugging Face paper, SoL-Pi, applies recursive self-improvement at the agent harness layer, matching Pi's EdgeBench performan...
Hugging Face Study Dissects What Actually Makes Coding Agents Work
A new empirical study from Hugging Face isolates planning, action space, and context management inside a fixed coding harness, eva...
PANORAMA Brings Pixel-Level Grounding to Detailed Image Captions
A new Hugging Face paper introduces PANORAMA, a vision-language model that grounds every referring phrase in a caption with pixel-...
New Memory-Bounding Techniques Let MoE Models Train at 1M Context
A new paper introduces four complementary methods that cap every memory peak in long-context Mixture-of-Experts training, enabling...
GPT-Policy Lets Robots Learn New Tasks In-Context via VLM Agents
A new Hugging Face paper introduces GPT-Policy, a general-agent framework that combines a context compiler, a commercial vision-la...