Research Papers
OpenAI Shares Frontier Model's Progress on Open Mathematics Problems
OpenAI has published new results from an internal frontier model tackling open problems in mathematics, releasing Lean proof forma...
Language Model Agents Rewrite Sella to Cut Molecular Relaxation Costs
A new Hugging Face paper introduces AutoSella, a family of molecular geometry optimizers discovered by language-model agents that...
Hugging Face Paper Introduces 'World Editing' as a New AI Capability, With Minecraft and Terraria Benchmark
A new paper formalizes world editing as intervening on an existing executable world while preserving unchanged properties, introdu...
SEER Turns Forecast Errors Into Smarter Event Retrieval for Time Series
A new closed-loop framework called SEER feeds forecasting errors back into event retrieval and causal knowledge, improving predict...
Hugging Face Paper Introduces Inverse Distillation Unlearning for One-Step Generative Models
A new framework called Inverse Distillation Unlearning (IDU) jointly distills multi-step diffusion and flow-matching teachers into...
New Post-Training Method Aligns Diffusion Language Models via Representation-Space MMD
Researchers propose a post-training objective that minimizes Maximum Mean Discrepancy between generated and reference distribution...
LoGRA Cuts LLM Reinforcement Learning Memory by Up to 45.7%
A new method called LoGRA compresses reinforcement learning signals into low-rank gradient sketches, cutting training memory by up...
OmniReasoning Targets the Blind Spot in Audio-Visual Joint Reasoning
A new Hugging Face paper introduces OmniReasoningBench, the OmniQA data engine, and a self-distillation method that together push...
Self-Generated Feedback Destabilizes Test-Time Training, New Study Finds
A new Hugging Face paper shows that when models learn from their own generated text during inference, prediction on independent hu...
Hugging Face Paper Proposes 3T Framework and Proactivity-Gym Testbed for Proactive LLM Agents
A new paper outlines three joint principles — Task Capability, Temporal Allocation, and Trust — for designing proactive LLM agents...
LMBuild Benchmark Tests Whether LLM Agents Can Build Objects That Actually Work
A new benchmark from Hugging Face researchers evaluates LLM agents on generating buildable, functional 3D structures rather than j...
MemAdapter Targets Memory-Induced Sycophancy in LLM Agents
A new framework called MemAdapter adaptively integrates retrieved memories so LLM agents can reason objectively instead of over-al...