Metraipolis › archive › dc_d0777fdb8751

World in World: Explore the World with World Models — arXiv:2609.11548

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.11548
License (as archived)
CC0-1.0
Archived
2026-10-16T06:35:43.670Z
SHA-256 of the archived text
5c62a6c75581bbf6eeaf66c567d95dbe638779a50a8181355d350dc73100ddcf
Stable URL of this page
https://metraipolis.com/documento/dc_d0777fdb8751

How to cite

World in World: Explore the World with World Models — arXiv:2609.11548 — arXiv (en), https://arxiv.org/abs/2609.11548 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d0777fdb8751 (Metraipolis archive dc_d0777fdb8751, sha256 5c62a6c75581bbf6eeaf66c567d95dbe638779a50a8181355d350dc73100ddcf)
BibTeX
@misc{dcd0777fdb8751,
  title = {World in World: Explore the World with World Models — arXiv:2609.11548},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.11548},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d0777fdb8751; sha256 5c62a6c75581bbf6eeaf66c567d95dbe638779a50a8181355d350dc73100ddcf},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.11548. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Autoregressive video world models enable interactive, long-horizon exploration, but flexible control remains challenging. Exploring a source video from new viewpoints requires the generated rollout to remain synchronised with the recorded event, place observed content in the requested view, plausibly complete newly exposed regions, and recover previously generated appearance on revisits. Existing methods typically address these requirements through task-specific modules or additional training. We present World in World, a training-free inference-time interface that converts heterogeneous control evidence into camera- and time-labelled clean visual states, which are read through the native self attention of a frozen causal video model. The evidence comprises source-video observations, target-view scene projections, geometry renderings that guide completion of newly exposed subject regions, and retrieved generated states beyond the rolling cache. Each evidence source carries token-level support and its own availability schedule. A correspondence router combines persistent point identities with geometry to establish token correspondences, guiding supported queries towards matching source-video tokens. Evidence-wise attention CFG (EWA) then independently regulates each auxiliary channel's additional contribution using attention responses from the same denoising forward pass. The shared interface supports camera-controlled rerendering, long-horizon revisiting, and human-motion transfer with the same frozen backbone. We evaluate World in World on camera-controlled video rerendering under diverse viewpoint changes, assessing perceptual quality, temporal consistency, and camera-following accuracy.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-11 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.11548 — 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 2026-10-16T06:06:28.793Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).