Metraipolis › archive › dc_7d9035c0dfc4

PhysicsLENS: Diagnosing Physical Property Blindness in Video Generation Models — arXiv:2610.01162

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.01162
License (as archived)
CC0-1.0
Archived
2026-10-06T19:12:43.205Z
SHA-256 of the archived text
ac1dd4053d20cdc72d0a22e64077f7eeb9cb7cc82f6d050b9157faaec28a59f3
Stable URL of this page
https://metraipolis.com/documento/dc_7d9035c0dfc4

How to cite

PhysicsLENS: Diagnosing Physical Property Blindness in Video Generation Models — arXiv:2610.01162 — arXiv (en), https://arxiv.org/abs/2610.01162 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7d9035c0dfc4 (Metraipolis archive dc_7d9035c0dfc4, sha256 ac1dd4053d20cdc72d0a22e64077f7eeb9cb7cc82f6d050b9157faaec28a59f3)
BibTeX
@misc{dc7d9035c0dfc4,
  title = {PhysicsLENS: Diagnosing Physical Property Blindness in Video Generation Models — arXiv:2610.01162},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.01162},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7d9035c0dfc4; sha256 ac1dd4053d20cdc72d0a22e64077f7eeb9cb7cc82f6d050b9157faaec28a59f3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01162. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reliable video world models could provide scalable predictive environments for robot learning, planning, and evaluation. However, generated robot videos can violate physical principles and complete tasks through physically implausible behavior, limiting their reliability for robot learning and planning. Current video-generation benchmarks exclude physics that are inherently hidden by visuals (e.g., weight, viscosity, friction). Due to this, video models are evaluated on the fidelity of physics, not the underlying accuracy of physics. We introduce PhysicsLENS, a dataset and benchmark for evaluating plausibility of physical properties grounded in robotics. PhysicsLENS uses matched scenario pairs that hold the same conditioning frame and task, while varying underlying physics in the scene description. Scenarios are curated from public robot video sources and annotated across seven physical domains: collision, gravity, momentum, friction, deformation, fluid, and causality. We evaluate across four video generation models, producing over 400 human-annotated labels. Results show that plausible-looking videos often ignore the stated property (34 of 47), and that stating the property lowers plausibility only slightly and not significantly.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2610.01162 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-06T17:37:36.747Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).