Metraipolis › archive › dc_8bc2eb56f20c

Revision, Not Restart: Revisable Visual Plans for Closed-Loop World-Action Models — arXiv:2609.35439

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.35439
License (as archived)
CC0-1.0
Archived
2026-10-07T23:55:13.041Z
SHA-256 of the archived text
9f2d033dddb89875d312d69982dd1bfa129a741eb54cf5e332f84439da5d9e6b
Stable URL of this page
https://metraipolis.com/documento/dc_8bc2eb56f20c

How to cite

Revision, Not Restart: Revisable Visual Plans for Closed-Loop World-Action Models — arXiv:2609.35439 — arXiv (en), https://arxiv.org/abs/2609.35439 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8bc2eb56f20c (Metraipolis archive dc_8bc2eb56f20c, sha256 9f2d033dddb89875d312d69982dd1bfa129a741eb54cf5e332f84439da5d9e6b)
BibTeX
@misc{dc8bc2eb56f20c,
  title = {Revision, Not Restart: Revisable Visual Plans for Closed-Loop World-Action Models — arXiv:2609.35439},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.35439},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8bc2eb56f20c; sha256 9f2d033dddb89875d312d69982dd1bfa129a741eb54cf5e332f84439da5d9e6b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35439. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

World-action models use predicted visual futures to condition robot actions, yet execution feedback can invalidate parts of a prediction while leaving its task structure useful. We propose Revisable Temporal Planning (RTP), which maintains the visual future as a persistent action condition and revises it after feedback. Its central mechanism is a learned revision bridge: it resumes an intermediate state saved during visual generation and adapts its continuation to current observations. Visual and action supervision connect this revision to subsequent control. Time-aware history supplies observed evidence, and an adaptive policy selects retention, bridge revision, or fresh replanning from new noise before decoding the next action. On RoboMME and RMBench, RTP achieves task-averaged success rates of 48.6% and 84.8%, respectively. Matched comparisons support learned continuation; estimated checkpoint-source and action-prefix effects are positive but less precisely resolved. These results connect feedback-driven visual-plan revision to closed-loop task performance. Project Page: this https URL

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.35439 — 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 2026-10-07T23:29:27.728Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).