Metraipolis › archive › dc_024cfa3c4ddb

A Safe Action Is Not Enough: Feasible-Future Decoding for Vision-Language-Action Policies — arXiv:2610.05166

Source
arXiv
Original
https://arxiv.org/abs/2610.05166
License (as archived)
CC0-1.0
Archived
2026-10-06T05:25:01.858Z
SHA-256 of the archived text
493850910061c850967727d1baa1ef3b787ec4d6887a2c2a00f3685a156626a7
Stable URL of this page
https://metraipolis.com/documento/dc_024cfa3c4ddb

How to cite

A Safe Action Is Not Enough: Feasible-Future Decoding for Vision-Language-Action Policies — arXiv:2610.05166 — arXiv, https://arxiv.org/abs/2610.05166 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_024cfa3c4ddb (Metraipolis archive dc_024cfa3c4ddb, sha256 493850910061c850967727d1baa1ef3b787ec4d6887a2c2a00f3685a156626a7)
BibTeX
@misc{dc024cfa3c4ddb,
  title = {A Safe Action Is Not Enough: Feasible-Future Decoding for Vision-Language-Action Policies — arXiv:2610.05166},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.05166},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_024cfa3c4ddb; sha256 493850910061c850967727d1baa1ef3b787ec4d6887a2c2a00f3685a156626a7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05166. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A safe action is not necessarily a viable one. Under a frozen vision-language-action (VLA) policy, an action can be likely and locally admissible yet leave no policy-supported route to safe task completion. We call this the feasibility-likelihood gap: likelihood ranks the current action, whereas feasibility depends on the futures that remain after it. We derive the exact next-block marginal of the history-conditioned policy-environment trajectory law restricted to safe task completion. The derivation exposes a candidate-dependent feasible-future mass with two roles: its support records whether safe completion remains possible under the frozen continuation process, and its magnitude measures how much weighted safe-completion mass is preserved. Exact evaluation is impractical online, so we develop a selective finite-candidate approximation, derive conditions for recovering the best retained viable candidate, and instantiate it as an alarm-triggered, training-free reranker. On Safety-CHORES, VICS-G lowers mean cumulative safety cost by 1.9%-57.5% across six settings while remaining within 2.5 percentage points of policy sampling in success and 0.82 steps in mean episode length. The resulting decoder is tied to an exact policy-relative safe-completion target, yet requires neither retraining of the base policy nor online trajectory rollouts.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2610.05166 — 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 2026-10-06T05:24:05.558Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).