Metraipolis › archive › dc_664be3462fc8

VLA-Scope: Shift-Aware Failure Prediction for Vision-Language-Action Models — arXiv:2609.21246

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.21246
License (as archived)
CC0-1.0
Archived
2026-10-10T01:07:54.500Z
SHA-256 of the archived text
679dbf754f2a2c012c42b117e11d2ac790c49c08f7678136a2af3aef8d75d801
Stable URL of this page
https://metraipolis.com/documento/dc_664be3462fc8

How to cite

VLA-Scope: Shift-Aware Failure Prediction for Vision-Language-Action Models — arXiv:2609.21246 — arXiv (en), https://arxiv.org/abs/2609.21246 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_664be3462fc8 (Metraipolis archive dc_664be3462fc8, sha256 679dbf754f2a2c012c42b117e11d2ac790c49c08f7678136a2af3aef8d75d801)
BibTeX
@misc{dc664be3462fc8,
  title = {VLA-Scope: Shift-Aware Failure Prediction for Vision-Language-Action Models — arXiv:2609.21246},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.21246},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_664be3462fc8; sha256 679dbf754f2a2c012c42b117e11d2ac790c49c08f7678136a2af3aef8d75d801},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.21246. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-language-action (VLA) models map visual observations and natural-language instructions to robotic actions, but distribution shifts can compromise their reliability. Because these models may still succeed under out-of-distribution (OOD) conditions, detecting OOD inputs alone is insufficient to predict execution failure. In this paper, we introduce VLA-Scope, a two-stage framework that combines input-shift characterization with execution history to predict failure during OOD rollouts. The first stage uses pooled image and language representations to detect OOD inputs and classify their shift categories. For inputs flagged as OOD, the second stage combines the predicted category, action-prefix features, and execution progress features. A logistic regression model shared across shift categories updates failure risk as execution proceeds. We evaluate the framework with OpenVLA on ten LIBERO-Spatial tasks using leave-one-group-out cross-validation. OOD detection achieves a ROC-AUC of 0.9454, and shift classification achieves 91% accuracy. Evaluated independently of the OOD gate on all 1,400 OOD rollouts, the failure predictor achieves a ROC-AUC of 0.8497 after 60 executed actions, compared with 0.7906 without execution progress features. It also achieves a higher ROC-AUC than the evaluated ActProbe and SAFE-MLP baselines. These results suggest that combining action features with temporally aggregated execution step representations improves failure prediction under input shifts.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-21 in cs.AI).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Systems and Control (eess.SY).
Pagina del lavoro: https://arxiv.org/abs/2609.21246 — 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 2026-10-10T00:23:42.520Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).