DataComp-VLM: Improved Open Datasets for Vision-Language Models
Paper • 2606.28551 • Published • 40
reading-buffer
-
-
SkillCoach: Self-Evolving Rubrics for Evaluating and Enhancing Agentic Skill-Use
Paper • 2607.01874 • Published • 18 -
PACE: A Proxy for Agentic Capability Evaluation
Paper • 2607.02032 • Published • 16 -
Measuring the Gap Between Human and LLM Research Ideas
Paper • 2607.01233 • Published • 10 -
Agents' Last Exam
Paper • 2606.05405 • Published • 374 -
WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces
Paper • 2606.09426 • Published • 104 -
OpenRath: Session-Centered Runtime State for Agent Systems
Paper • 2606.19409 • Published • 77 -
A Matter of TASTE: Improving Coverage and Difficulty of Agent Benchmarks
Paper • 2605.28556 • Published • 73 -
Code as Agent Harness
Paper • 2605.18747 • Published • 224 -
From Context to Skills: Can Language Models Learn from Context Skillfully?
Paper • 2604.27660 • Published • 171 -
Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability
Paper • 2604.06628 • Published • 329 -
The Latent Space: Foundation, Evolution, Mechanism, Ability, and Outlook
Paper • 2604.02029 • Published • 152 -
Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe
Paper • 2604.13016 • Published • 113 -
Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence
Paper • 2604.18292 • Published • 88 -
GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization
Paper • 2601.05242 • Published • 235 -
Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation
Paper • 2602.12125 • Published • 68