Who Takes the Lead? Model Ability and Task-Conditioned Collaboration with Humans Paper • 2610.27148 • Published 8 days ago • 12
From Failures to Learnable Worlds: Policy-Conditioned Environment Generation for Navigation Agents Paper • 2610.19420 • Published 13 days ago • 10
Questioning the Questions: Sustaining Self-Evolution in Reasoning Models Paper • 2610.04299 • Published 24 days ago • 86
OTRetarget: Joint Robot and Object Motion Retargeting via Optimal Transport Paper • 2609.36602 • Published 28 days ago • 8
Can MiniMax-H3 Reason About the Physical World? An Evaluation of Omni-Modal Generative Model Paper • 2609.18323 • Published Sep 16 • 96
Agentic Visual Generation: From Generative Models to Agentic Control Paper • 2609.06758 • Published Sep 6 • 35