Metraipolis › archive › dc_d67502c91058

Physis-Lang: Self-Evolving Language as a Physical Representation for Video World Model — arXiv:2609.40358

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.40358
License (as archived)
CC0-1.0
Archived
2026-10-07T14:46:42.552Z
SHA-256 of the archived text
965d6a8de4803ed78d402a6e5a04b3a0519be7168ae0a31bcb655eef2d498883
Stable URL of this page
https://metraipolis.com/documento/dc_d67502c91058

How to cite

Physis-Lang: Self-Evolving Language as a Physical Representation for Video World Model — arXiv:2609.40358 — arXiv (en), https://arxiv.org/abs/2609.40358 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d67502c91058 (Metraipolis archive dc_d67502c91058, sha256 965d6a8de4803ed78d402a6e5a04b3a0519be7168ae0a31bcb655eef2d498883)
BibTeX
@misc{dcd67502c91058,
  title = {Physis-Lang: Self-Evolving Language as a Physical Representation for Video World Model — arXiv:2609.40358},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.40358},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d67502c91058; sha256 965d6a8de4803ed78d402a6e5a04b3a0519be7168ae0a31bcb655eef2d498883},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.40358. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Video world models are expected to predict how the physical world evolves, yet they often produce visually plausible videos that violate basic physical principles. Existing approaches commonly assume that natural language is insufficient to represent the physical knowledge required for reliable generation, and therefore introduce additional visual, latent, numerical, or planning-based signals. We revisit this assumption and introduce Physis-Lang, a self-evolving framework that treats physical language as a shared and optimizable representation across data curation, model training, and video generation. Physis-Lang represents physical processes through language that describes their relevant entities, causes, interactions, governing principles, temporal evolution, and effects. To improve this representation, we construct PhysCapBench, which decomposes physical processes into atomic assertions and evaluates captions using recall and precision. An agentic loop iteratively analyzes assertion-level errors and refines the instruction used to produce physical captions. Physis-Lang further converts model deficiencies into textual descriptions and uses language-guided retrieval to identify visually diverse videos that cover missing physical processes. Experiments on four widely used physical video benchmarks with Wan and Cosmos backbones demonstrate consistent improvements in physical plausibility. Notably, starting from open-source Cosmos3-Nano backbones, our Physis-Lang-enhanced models surpass the leading proprietary Veo 3.1 model.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.40358 — 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 2026-10-07T14:19:00.870Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).