Metraipolis › archive › dc_36832fb22dd6

WorldAlign: Decoupled 4D Reward for World-Consistent Video Generation — arXiv:2610.12382

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.12382
License (as archived)
CC0-1.0
Archived
2026-10-09T06:30:57.925Z
SHA-256 of the archived text
47703ec1b0bcc57313563410fdbbf0ea576d159492d71d108b32f3a7719dcbe2
Stable URL of this page
https://metraipolis.com/documento/dc_36832fb22dd6

How to cite

WorldAlign: Decoupled 4D Reward for World-Consistent Video Generation — arXiv:2610.12382 — arXiv (en), https://arxiv.org/abs/2610.12382 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_36832fb22dd6 (Metraipolis archive dc_36832fb22dd6, sha256 47703ec1b0bcc57313563410fdbbf0ea576d159492d71d108b32f3a7719dcbe2)
BibTeX
@misc{dc36832fb22dd6,
  title = {WorldAlign: Decoupled 4D Reward for World-Consistent Video Generation — arXiv:2610.12382},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.12382},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_36832fb22dd6; sha256 47703ec1b0bcc57313563410fdbbf0ea576d159492d71d108b32f3a7719dcbe2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.12382. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Faithful visual world simulation requires generated videos to maintain 4D world consistency, encompassing both static and dynamic consistency. Static consistency requires coherent 3D structure in static environments across viewpoints, while dynamic consistency requires plausible subject motion and consistent appearance over time. Geometry-aware post-training offers a promising way to improve world consistency. However, existing methods often rely on a static-scene assumption. Even those that accommodate dynamic scenes struggle to provide reliable static-consistency feedback, while dynamic consistency is often overlooked or inadequately assessed. To address these limitations, we introduce WorldAlign, a decoupled 4D reward framework that semantically separates static regions and dynamic subjects and provides feedback by aligning each with a world prior suited to its assumptions. For static regions, WorldAlign aligns static geometry with a geometric world prior through semantically guided masked reprojection, enabling more reliable static-consistency evaluation; an auxiliary camera-motion reward discourages nearly static solutions. For dynamic subjects, WorldAlign uses a strong vision-language model (VLM) as a dynamic world prior and constructs a VLM-as-a-judge reward based on sample-specific checklists that assess dynamicity, physical plausibility, shape, and texture consistency. This decoupled design enables more effective online post-training without requiring human preference annotations. Across two pretrained image-to-video generators, Wan2.1 and Wan2.2, WorldAlign jointly improves static and dynamic consistency over existing methods without suppressing overall or subject motion. These results support decoupled world-prior alignment for more faithful visual world simulation. Project page: this https URL .

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.12382 — 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 2026-10-09T05:55:07.699Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).