Metraipolis › archive › dc_49c44e6750fa

SafeStage: Evaluating Safety Before, During, and After Vision-Language-Conditioned Robot Manipulation — arXiv:2609.21223

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.21223
License (as archived)
CC0-1.0
Archived
2026-10-10T02:30:03.944Z
SHA-256 of the archived text
0686f089bcb4d55b3add93ab3402202bde8908fc7f28a8ab917dd816a20c858d
Stable URL of this page
https://metraipolis.com/documento/dc_49c44e6750fa

How to cite

SafeStage: Evaluating Safety Before, During, and After Vision-Language-Conditioned Robot Manipulation — arXiv:2609.21223 — arXiv (en), https://arxiv.org/abs/2609.21223 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_49c44e6750fa (Metraipolis archive dc_49c44e6750fa, sha256 0686f089bcb4d55b3add93ab3402202bde8908fc7f28a8ab917dd816a20c858d)
BibTeX
@misc{dc49c44e6750fa,
  title = {SafeStage: Evaluating Safety Before, During, and After Vision-Language-Conditioned Robot Manipulation — arXiv:2609.21223},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.21223},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_49c44e6750fa; sha256 0686f089bcb4d55b3add93ab3402202bde8908fc7f28a8ab917dd816a20c858d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.21223. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-language-conditioned robot policies integrate perception, language understanding, and control for general-purpose manipulation. However, existing evaluations often focus on task success, isolated physical constraints, semantic refusal, or realized physical damage, providing limited insight into where safety fails during closed-loop manipulation. We introduce SafeStage, a lifecycle-structured benchmark for evaluating manipulation safety before, during, and after task execution. SafeStage contains 97 purpose-built risk scenarios organized into three stages. Initial-State Hazards captures safety-relevant relations that must be resolved before manipulating the target. Execution-Time Safety evaluates unsafe contacts, trajectories, region entries, and object interactions during execution. Final-State Hazards capture unstable or otherwise unsafe conditions remaining after nominal task completion. The benchmark evaluates realized interactions using event-based and state-based checks and reports native task success independently from stage-specific safety outcomes. We evaluate representative direct-action Vision-Language-Action (VLA) policies and policies with world-model-based policies under a common closed-loop protocol. Our results demonstrate that nominal task completion frequently coexists with safety violations and that different policies exhibit distinct failure profiles across the three stages. By separating task success from safety and localizing when violations occur, SafeStage provides a unified diagnostic testbed for evaluating and improving vision-language-conditioned robot manipulation policies.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-21 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.21223 — 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 2026-10-10T02:09:55.324Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).