Metraipolis › archive › dc_d2ef093ddc90

Confidence-Modulated Safety Filtering for End-to-End Policies — arXiv:2610.40922

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.40922
License (as archived)
CC0-1.0
Archived
2026-11-02T08:03:45.760Z
SHA-256 of the archived text
77b36fa7cdbe462f40c28d7453b4bf57e34ccbe09cbc123914b49b3642467d5a
Stable URL of this page
https://metraipolis.com/documento/dc_d2ef093ddc90

How to cite

Confidence-Modulated Safety Filtering for End-to-End Policies — arXiv:2610.40922 — arXiv (en), https://arxiv.org/abs/2610.40922 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d2ef093ddc90 (Metraipolis archive dc_d2ef093ddc90, sha256 77b36fa7cdbe462f40c28d7453b4bf57e34ccbe09cbc123914b49b3642467d5a)
BibTeX
@misc{dcd2ef093ddc90,
  title = {Confidence-Modulated Safety Filtering for End-to-End Policies — arXiv:2610.40922},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.40922},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d2ef093ddc90; sha256 77b36fa7cdbe462f40c28d7453b4bf57e34ccbe09cbc123914b49b3642467d5a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.40922. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

End-to-end policies, from task-specific controllers to robot foundation models, learn complex behavior from raw observations but are not guaranteed to remain safe, so they are often wrapped in a safety filter. The resulting system operates under two domains: the filter's operational design domain (ODD), where its assumptions hold and safety is certified, and the policy's competence domain, where it is reliable. Filters typically fix their conservativeness regardless of the policy's competence, trading performance inside the ODD against margin beyond it. We ask whether the policy's confidence should set that level, and propose a filter-agnostic pipeline that modulates it at runtime: inside the ODD, where every setting is safe, confidence buys performance; beyond it, low confidence adds margin, which helps when it flags where the filter's model fails. Preliminary F1TENTH simulations, where this holds, show the modulated filter combining the strengths of permissive and conservative filters.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.RO, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Systems and Control (eess.SY).
Pagina del lavoro: https://arxiv.org/abs/2610.40922 — 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 2026-11-02T07:44:15.689Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).