Causal Physics Steering in Video World Models via Concept Activation Vectors 事件

Causal Physics Steering in Video World Models via Concept Activation Vectors arXiv:2605.24322v1 Announce Type: new Abstract: Video world models learn representations of physical dynamics, but controlling their physical expectations at inference time remains an open problem. Recent interpretability work identified a Physics Emergence Zone (PEZ), a group of middle transformer layers in VideoMAE where physical plausibility is represented separately from other visual features. However, it remained