Metraipolis › archive › dc_f65fef9f81df

Precise Editing and Flexible Referencing for Interactable Worlds — arXiv:2609.34470

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.34470
License (as archived)
CC0-1.0
Archived
2026-10-07T21:02:38.331Z
SHA-256 of the archived text
1fc97380d25834630fd1ca08db86c33459cc97059529cdef95a7083b2eb30206
Stable URL of this page
https://metraipolis.com/documento/dc_f65fef9f81df

How to cite

Precise Editing and Flexible Referencing for Interactable Worlds — arXiv:2609.34470 — arXiv (en), https://arxiv.org/abs/2609.34470 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f65fef9f81df (Metraipolis archive dc_f65fef9f81df, sha256 1fc97380d25834630fd1ca08db86c33459cc97059529cdef95a7083b2eb30206)
BibTeX
@misc{dcf65fef9f81df,
  title = {Precise Editing and Flexible Referencing for Interactable Worlds — arXiv:2609.34470},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.34470},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f65fef9f81df; sha256 1fc97380d25834630fd1ca08db86c33459cc97059529cdef95a7083b2eb30206},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34470. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present EditWorld, a video world model for precise editing and flexible referencing in interactable worlds. Existing video world models primarily focus on navigation, letting users explore generated worlds but offering limited control over how existing world content is modified. EditWorld extends world modeling from exploration to precise modification by streaming editing instructions and reference images during autoregressive generation. To support these capabilities, EditWorld introduces Gated Causal Attention for temporally varying editing conditions and reference images, together with a Sparse Context mechanism that maintains a bounded historical context for long-horizon inference. We further adopt joint autoregressive and bidirectional training with annealed self-resampling, and construct a dedicated data synthesis and annotation pipeline that provides supervision for world editing. We also present WBench-Editing to systematically evaluate streaming world editing capabilities. EditWorld achieves the best overall performance on WBench-Editing with an overall score of 73.8 and an editing score of 80.0, substantially outperforming existing methods on editing-related metrics. this https URL

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.34470 — 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 2026-10-07T20:29:03.005Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).