Metraipolis › archive › dc_0786d834d359

Natural State-Prediction Accuracy can Hide Weak Controlled Responsiveness in VLA Readouts — arXiv:2609.34684

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.34684
License (as archived)
CC0-1.0
Archived
2026-10-07T22:40:41.864Z
SHA-256 of the archived text
901c3265cb0344dc58324c104f94366801bc49f8cabaf507140c5c0a79f3c835
Stable URL of this page
https://metraipolis.com/documento/dc_0786d834d359

How to cite

Natural State-Prediction Accuracy can Hide Weak Controlled Responsiveness in VLA Readouts — arXiv:2609.34684 — arXiv (en), https://arxiv.org/abs/2609.34684 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0786d834d359 (Metraipolis archive dc_0786d834d359, sha256 901c3265cb0344dc58324c104f94366801bc49f8cabaf507140c5c0a79f3c835)
BibTeX
@misc{dc0786d834d359,
  title = {Natural State-Prediction Accuracy can Hide Weak Controlled Responsiveness in VLA Readouts — arXiv:2609.34684},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.34684},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0786d834d359; sha256 901c3265cb0344dc58324c104f94366801bc49f8cabaf507140c5c0a79f3c835},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34684. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Accurately decoding object states from the internal representations of vision-language-action (VLA) models does not establish that the predictions respond faithfully to changes in the target physical state. In natural observations, object state, robot configuration, occlusion, and task progress vary together, allowing contextual cues to contribute to prediction. In this paper, we introduce an evaluation framework that separates prediction accuracy, target-state responsiveness, and context stability using physically validated observations that cross target coordinates with robot contexts. We demonstrate that high natural-trajectory accuracy can coexist with weak controlled target-state responsiveness in fixed representation-readout pairs. Comparisons and interventions involving representations, readouts, and training data show that the three properties provide distinct diagnostic information. Furthermore, adding responsiveness and context sensitivity to a failure predictor based on initial state error and physical variables reduces policy-failure prediction error on new initializations relative to the specified baseline while same-observation controlled MAE is also informative. These findings motivate evaluating target-state responsiveness and context stability alongside natural prediction accuracy, and examining their relationship to actual policy behavior and task outcomes.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.34684 — 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 2026-10-07T22:14:28.853Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).