Metraipolis › archive › dc_25fe759379eb
Accelerating VLA Inference with Perturbation Informed Execution — arXiv:2610.19656
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.19656
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T21:57:55.818Z
- SHA-256 of the archived text
9d4bbf5cfcdd20239f51e1f9618d40d027b6436f9caa2d8ce336e78ae578f5bd- Stable URL of this page
https://metraipolis.com/documento/dc_25fe759379eb
How to cite
Accelerating VLA Inference with Perturbation Informed Execution — arXiv:2610.19656 — arXiv (en), https://arxiv.org/abs/2610.19656 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_25fe759379eb (Metraipolis archive dc_25fe759379eb, sha256 9d4bbf5cfcdd20239f51e1f9618d40d027b6436f9caa2d8ce336e78ae578f5bd)
BibTeX
@misc{dc25fe759379eb,
title = {Accelerating VLA Inference with Perturbation Informed Execution — arXiv:2610.19656},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.19656},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_25fe759379eb; sha256 9d4bbf5cfcdd20239f51e1f9618d40d027b6436f9caa2d8ce336e78ae578f5bd},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19656. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Action-chunking vision-language-action (VLA) models predict sequences of future robot actions. Under fixed-horizon execution, the robot executes a fixedlength prefix of each predicted chunk before invoking the policy again, leaving the remaining predictions unused. For a given number of executed actions, short execution horizons require more policy calls. Extending the execution horizon amortizes policy inference cost over more actions. However, uniformly applying long horizons across episodes can limit opportunities for precise corrections. To address this limitation, we propose PICES (Perturbation-Informed Chunk and Execution Horizon Selection), a training-free method that jointly selects an action chunk and the number of actions executed before the next policy call. Within a single policy call, the action expert generates a batch of action chunks from multiple initial noise samples and their locally perturbed counterparts, using the same conditioning features across all samples. PICES exploits differences between corresponding actions in each nominal chunk and its counterpart to determine candidate-specific execution horizons and select a nominal chunk for execution. Experiments on simulation benchmarks and real-world manipulation tasks demonstrate that PICES achieves speedup by up to 2.3x in wall-clock policy inference time on Jetson Orin and Thor relative to fixed horizon baselines while improving the success rate. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.RO, nuovi lavori del giorno). Categorie arXiv: Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2610.19656 — 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 2026-10-15T21:38:44.472Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).