Physics-Editable World Models: Data, Interventions, and Evaluation for Controllable Simulation

Abstract

Physics-Editable World Models now turns on the ability to balance performance with evidence quality, resource limits, and transfer across settings. This technical review examines distinguishing visually plausible editing from interventions that preserve physical consistency. The comparison integrates 1 focal paper with 11 independently retrieved publications verified through persistent DOI or publisher records. The analysis is organized around dataset design, physical factors, counterfactual editing, temporal consistency, and evaluation. A central precaution is not to treat results from heterogeneous studies as exchangeable, the review compares problem formulation, design assumptions, and transfer boundary. Across the literature, the comparison suggests that advances in physics-editable world models become credible when representation, objective, and evaluation protocol are evaluated together and when uncertainty about distribution shift is reported explicitly. The synthesis ties method selection to use-case risk and reveals common limits on generalization, and proposes a research agenda centered on transparent baselines, stress testing, and reproducible evidence.