Metraipolis › archive › dc_bdc1caef0941

CereVLA: Cerebellum-Inspired Consequence-Aware Residual Governance for Efficient Vision-Language-Action Execution — arXiv:2609.27468

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.27468
License (as archived): CC0-1.0
Archived: 2026-10-08T12:34:52.749Z
SHA-256 of the archived text: 5371ac37766abee73c43d1ed7377b1d9be3c9cd921d986f58761b2bc033076b5
Stable URL of this page: https://metraipolis.com/documento/dc_bdc1caef0941

How to cite

CereVLA: Cerebellum-Inspired Consequence-Aware Residual Governance for Efficient Vision-Language-Action Execution — arXiv:2609.27468 — arXiv (en), https://arxiv.org/abs/2609.27468 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bdc1caef0941 (Metraipolis archive dc_bdc1caef0941, sha256 5371ac37766abee73c43d1ed7377b1d9be3c9cd921d986f58761b2bc033076b5)

BibTeX

@misc{dcbdc1caef0941,
  title = {CereVLA: Cerebellum-Inspired Consequence-Aware Residual Governance for Efficient Vision-Language-Action Execution — arXiv:2609.27468},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.27468},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bdc1caef0941; sha256 5371ac37766abee73c43d1ed7377b1d9be3c9cd921d986f58761b2bc033076b5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.27468. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Action-chunked vision-language-action (VLA) policies improve inference efficiency, but limited feedback within committed action chunks can lead to accumulated execution errors. Residual adaptation can correct such deviations without retraining the VLA; however, existing corrections are typically optimized for reference-action consistency without explicitly considering their downstream consequences. To address this limitation, we present Cerebellum-Inspired Consequence-Aware Residual Governance (CereVLA), a unified framework that integrates lightweight residual refinement and predictive consequence evaluation into frozen VLA execution. Corrective actions are first generated by flow-based residual refinement, and their short- and interval-horizon consequences are then evaluated by a recurrent state-space model and a history-aware classifier. Residual corrections predicted to be unfavorable are selectively suppressed by a lightweight governor. Comparisons with state-of-the-art methods on LIBERO-10 and LIBERO-GOAL demonstrate the effectiveness of CereVLA. On SO-101, CereVLA increases task success from 57.5% to 90.0% and reduces mean control steps by 19.6% among successful trials, relative to the frozen SmolVLA baseline.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-24 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Systems and Control (eess.SY).
Pagina del lavoro: https://arxiv.org/abs/2609.27468 — 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 2026-10-08T12:12:35.687Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).