Metraipolis › archive › dc_f886cefde145
World Embedding Benchmark — arXiv:2610.03632
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.03632
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-05T22:52:02.483Z
- SHA-256 of the archived text
ba414bebd791b3cb8726928e0f432efa3406578e86704e1ce8ac622231663d5d- Stable URL of this page
https://metraipolis.com/documento/dc_f886cefde145
How to cite
World Embedding Benchmark — arXiv:2610.03632 — arXiv, https://arxiv.org/abs/2610.03632 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f886cefde145 (Metraipolis archive dc_f886cefde145, sha256 ba414bebd791b3cb8726928e0f432efa3406578e86704e1ce8ac622231663d5d)
BibTeX
@misc{dcf886cefde145,
title = {World Embedding Benchmark — arXiv:2610.03632},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.03632},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f886cefde145; sha256 ba414bebd791b3cb8726928e0f432efa3406578e86704e1ce8ac622231663d5d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.03632. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Physical fidelity has received increasing attention in world models and video generation, yet how video representations encode physical information remains less understood. We introduce the World Embedding Benchmark, comprising 8,000 controlled simulation cases from 80 families spanning fluid mechanics, solid mechanics, dynamics, and optics & electromagnetism. Each case pairs a rendered video with simulation-derived physical annotations, supporting three complementary tasks: text-video retrieval, physical-property regression, and multiple-choice video-description pair classification. We use these tasks to distinguish cross-modal physical alignment from the recoverability of quantitative physical information. Evaluated pre-trained omnimodal embedding models show weak retrieval and near-chance within-family pair classification, while lightweight probes recover useful physical information from frozen video embeddings. Continual contrastive training with physics-specific video-text pairs improves retrieval and pair classification but degrades physical-property regression, revealing a trade-off between alignment and quantitative information recoverability. Finally, we use the embeddings to retrieve reference videos for retrieval-augmented generation with MiniMax-H3. Retrieved references improve the physical fidelity of generated videos, with stronger retrieval models yielding larger gains in our experiments. Together, these findings highlight the need to evaluate physical alignment and property recoverability jointly, and demonstrate the utility of physical representations for improving video generation. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.03632 — 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 2026-10-05T22:46:10.633Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).