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.
Stateful Harness Lifts World Model Accuracy on Hidden State from 32% to 92%
Researchers split rule execution from video rendering: Stateful Harness with a frozen MiniMax-H3 gets the correct outcome in 92% of WorldState-Bench pairs, versus…
ActionRoPE Turns Player Actions into Coordinates and Cuts Scene Drift by 66%
ActionRoPE controls game world models through world coordinates with no new parameters, cutting revisit drift by 66% compared with the best baseline.
Heartbeat Thinking Teaches Voice Agents to Work While You're Still Talking: Latency Down 36.4%
A new scheme for voice agents starts reasoning and tool calls mid-speech. On FDBench v3, tool coverage rose by 16 percentage points and response latency fell by…
Moonshot releases open Kimi K2 Thinking: 300 tool calls in a row
MTEdit: 500 thousand editing trajectories teach models not to lose quality over 8 steps
Conditional Program shrinks DeepSeek's n-gram memory from 203 to 51 GB at 2 bits
RoboHarness: A Coding Agent on Qwen3.8-Flash-Next Beats VLA Models in BEHAVIOR
New Survey Maps AI in Humanitarian Work: Six Tasks, From Weather to Refugees
A survey on Hugging Face Daily Papers describes AI as a tool for six humanitarian tasks: weather forecasting, energy, forests, poverty, post-disaster damage and…
A Frozen 16 MB Vocabulary Matches a Trained One in Small LLMs
Researchers replaced 168M embedding parameters in a Gemma-like model with a frozen 16 MB table and did not lose to the trainable version on three of four benchmarks.
Data Rewriting Beats DP-SGD on the Privacy-Quality Trade-Off
Kolab: 115 Hours of Recordings Show That AI Agents Don't Understand Their Human Colleagues
DialFork measured procedural hallucinations in agents across 36,278 dialogues
AlphaBet trains an LLM agent to forecast events: Brier score improves 14.4%
REASON teaches a model to think in two loops and saves about half the compute
REASON splits inference into a fast loop and a slow loop: the system gathers extra data only where uncertainty remains, and spends about half the compute.
KANLab: A TensorFlow Playground Fork That Shows KAN Training in the Browser
Lukas Niehaus and colleagues released KANLab: a small Kolmogorov–Arnold network trains in the browser, and every B-spline curve can be viewed and edited by hand.