frontier models
The Video Critic Says When Physics Breaks
PhysTrace compares history-conditioned expectations with observed windows, localizing physical inconsistencies in time without task-specific probes.
Summary
PhysTrace compares history-conditioned expectations with observed windows, localizing physical inconsistencies in time without task-specific probes.
Video models used as world models need more than a global plausible-or-impossible score. PhysTrace asks a multimodal model to form history-conditioned physical expectations, aligns them with later video windows and marks where observations conflict. The authors add violation intervals to an existing benchmark and report hit rates up to 70.7 percent across physical principles, plus stronger overall possible-versus-impossible discrimination than the compared methods. Applied to reconstructed model trajectories, it exposes attribute- and noise-dependent violation patterns. This is an interpretability evaluation, not proof that the model understands physics.
Why it matters
PhysTrace compares history-conditioned expectations with observed windows, localizing physical inconsistencies in time without task-specific probes.
Limits and context
- This is an interpretability evaluation, not proof that the model understands physics.
Key claims
PhysTrace compares history-conditioned expectations with observed windows, localizing physical inconsistencies in time without task-specific probes.
Qualification: This is an interpretability evaluation, not proof that the model understands physics.
Evidence: source-2026-10-24-011
Sources
- arXiv preprint 2610.31429, original v1arXiv · primary research
Corrections
No corrections have been recorded for this story.