Metraipolis › archive › dc_83fef2cf492c

ReSync: Re-Aligning the Two Clocks of Asynchronous World-Action Models — arXiv:2609.33944

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33944
License (as archived)
CC0-1.0
Archived
2026-10-08T00:17:15.071Z
SHA-256 of the archived text
8e48433197412d8f9dd95166786573211fcdcf70fab87730023b89d7c5cdcf76
Stable URL of this page
https://metraipolis.com/documento/dc_83fef2cf492c

How to cite

ReSync: Re-Aligning the Two Clocks of Asynchronous World-Action Models — arXiv:2609.33944 — arXiv (en), https://arxiv.org/abs/2609.33944 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_83fef2cf492c (Metraipolis archive dc_83fef2cf492c, sha256 8e48433197412d8f9dd95166786573211fcdcf70fab87730023b89d7c5cdcf76)
BibTeX
@misc{dc83fef2cf492c,
  title = {ReSync: Re-Aligning the Two Clocks of Asynchronous World-Action Models — arXiv:2609.33944},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33944},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_83fef2cf492c; sha256 8e48433197412d8f9dd95166786573211fcdcf70fab87730023b89d7c5cdcf76},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33944. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Jointly generating future video and actions has become a standard recipe for world-action models, and the strongest systems denoise the two streams on separate schedules: actions are decoded in few steps so control stays fast, while the video stream runs longer to keep the predicted future sharp. The design is deliberate, but it leaves the two streams on different clocks, and an action can become executable while the future that should justify it is still largely unresolved. We formalize this as a two-clock view of asynchronous inference and introduce the commitment-evidence gap, a quantity read directly from a model's own sampling schedule rather than measured by search. The gap is predictive: as it widens, candidate utility becomes harder to identify and extra candidate sampling buys less, while advancing the world stream buys more, and the two cross. Spending more world computation is therefore not simply better. The useful interval is closed at both ends, and both ends can be read off the schedule before any rollout. ReSync places the computation inside it: hold the action state, advance only the world within the supported window, then resume native denoising. No parameters change and no candidates are compared. On a frozen paired RoboCasa panel this improves success by 4.48 points, while an equal-compute control that waits without advancing the world does not move, and the same rule transfers to a second benchmark and a second backbone without retuning.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.33944 — 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 2026-10-07T23:52:16.505Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).