Metraipolis › archive › dc_f4c953e782ca

CalVLA: Shaping Native Confidence in Vision-Language-Action Policies with Outcome Supervision — arXiv:2611.00394

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.00394
License (as archived)
CC0-1.0
Archived
2026-11-03T10:41:49.782Z
SHA-256 of the archived text
feea7503135d62634e4e698d92c881a640db497c14ffa64868edc2a8a9a46b76
Stable URL of this page
https://metraipolis.com/documento/dc_f4c953e782ca

How to cite

CalVLA: Shaping Native Confidence in Vision-Language-Action Policies with Outcome Supervision — arXiv:2611.00394 — arXiv (en), https://arxiv.org/abs/2611.00394 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f4c953e782ca (Metraipolis archive dc_f4c953e782ca, sha256 feea7503135d62634e4e698d92c881a640db497c14ffa64868edc2a8a9a46b76)
BibTeX
@misc{dcf4c953e782ca,
  title = {CalVLA: Shaping Native Confidence in Vision-Language-Action Policies with Outcome Supervision — arXiv:2611.00394},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.00394},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f4c953e782ca; sha256 feea7503135d62634e4e698d92c881a640db497c14ffa64868edc2a8a9a46b76},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00394. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Can a vision-language-action (VLA) policy signal likely task failure from its own action probabilities? We introduce CalVLA, which shapes native action-token confidence using successful and failed rollouts and an Action-Confidence Alignment (ACal) squared margin-ranking loss. Low-rank adapters update OpenVLA-7B without an auxiliary failure detector. Observations are paired within tasks at similar normalized elapsed times, not necessarily equivalent physical progress. On a held-out test split from 500 LIBERO-10 configurations, adapted confidence achieves a joint failure-discrimination AUROC of $0.7392\pm0.0480$ over three seeds, averaging scores at 130 and 260 steps. Frozen-policy probes reach 0.6146--0.6436, but superiority over the linear probe is statistically inconclusive. For the same confidence summary, AUROC rises from 0.4485 before adaptation to 0.7327 after adaptation. Mean task success decreases from 58% to 55%. These results show improved failure discrimination, not improved control or established safety benefits.

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/2611.00394 — 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 2026-11-03T10:39:50.160Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).