MLS-Bench: A Holistic and Rigorous Assessment of AI Systems on Building Better AI Paper • 2605.08678 • Published 11 days ago • 8
Post-Trained MoE Can Skip Half Experts via Self-Distillation Paper • 2605.18643 • Published 2 days ago • 27
VideoAgentTrek: Computer Use Pretraining from Unlabeled Videos Paper • 2510.19488 • Published Oct 22, 2025 • 22
Scaling Computer-Use Grounding via User Interface Decomposition and Synthesis Paper • 2505.13227 • Published May 19, 2025 • 46