cs.CVNov 19, 2026

Physical Laws without Physics Loss? Probing Learned Dynamics in Video World Models through Forced Response

Authors: Yutao Li, Chen Wang, Wangmeng Zuo, Hui Li

Organizations: Harbin Institute of Technology

Abstract

Do video world models learn physical laws without a physics loss, or merely imitate motion? Overall motion and appearance scores cannot settle it. We investigate this through nonlinear forced response: an excitation shakes the support of a suspended object. We introduce dual visual-numeric conditioning and adapt six image-to-video backbones to predict this motion using only the flow-matching loss. Wan predicts motion accurately and generalizes to unseen excitations, objects, poses, and scenes. Counterfactual tests suggest capabilities beyond replaying training videos. But what laws have they learned? We design three tiers of tests using Taylor expansions of the dynamics and geometry. All six reproduce the linear response, including resonance; five capture higher-order dynamics, including nonlinear softening. Remarkably, softening is barely present in training: only 0.74% of the swing's squared displacement. Extensions to a double pendulum and a rocking block show sustained dynamics are learned more faithfully than brief impacts. Yet learned dynamics may fail to appear when state estimation or propagation goes wrong. We turn these tests and findings into strategies to improve video training and generation.

Figures & tables

Explore similar work

CardsList
  1. PhysResponse: Probing the Physics of Video Generation with Controlled Interventions

    Nov 18, 2026Yuhan Li, Shuran Zhang, Haixin Tian +12Physical Consistency in Video GenerationPhysical Reasoning in Video

  2. When Does Latent Prediction Error Support Physical Surprise Claims in Video World Models?

    Nov 1, 2026Anton Hristov, Aleksandar Slavov, Vasil VasilevVideo World ModelsWorld Model Evaluation