DataComp-VLM: Improved Open Datasets for Vision-Language Models Paper • 2606.28551 • Published 22 days ago • 52
The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning Paper • 2606.29526 • Published 20 days ago • 167
PixelEyes: Decoupling Perception and Reasoning for Pinpoint Visual Evidence Seeking Paper • 2607.00115 • Published 18 days ago • 12
Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players Paper • 2605.28816 • Published May 27 • 432
TerminalWorld: Benchmarking Agents on Real-World Terminal Tasks Paper • 2605.22535 • Published May 21 • 11