Preprint
Aug 2026
PhysMind: From Video to Executable Worlds for Training-Free Physical Reasoning
PhysMind is introduced, a training-free agentic framework that constructs one reusable, question-agnostic executable world per video that fits analytic continuous-time dynamics and latent physical parameters without unrolling a time-stepped simulator.
Chen Yang, Shenxiang Zeng, Haoyang Zhao et al.
· 1 citation