Metraipolis › archive › dc_c3bbdb41b671

VLALagBench: Understanding When and Why VLA Inference Latency Matters — arXiv:2611.03103

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.03103
License (as archived): CC0-1.0
Archived: 2026-11-03T18:57:26.145Z
SHA-256 of the archived text: 2637510f726ce559dcab5919098b9cdd9ba1143612960790eaa2a95f6aec6ff1
Stable URL of this page: https://metraipolis.com/documento/dc_c3bbdb41b671

How to cite

VLALagBench: Understanding When and Why VLA Inference Latency Matters — arXiv:2611.03103 — arXiv (en), https://arxiv.org/abs/2611.03103 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c3bbdb41b671 (Metraipolis archive dc_c3bbdb41b671, sha256 2637510f726ce559dcab5919098b9cdd9ba1143612960790eaa2a95f6aec6ff1)

BibTeX

@misc{dcc3bbdb41b671,
  title = {VLALagBench: Understanding When and Why VLA Inference Latency Matters — arXiv:2611.03103},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03103},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c3bbdb41b671; sha256 2637510f726ce559dcab5919098b9cdd9ba1143612960790eaa2a95f6aec6ff1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03103. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-language-action (VLA) models offer a promising path toward general-purpose robot manipulation. While inference latency can affect task success in a constantly evolving physical world, the relationship between latency and task success rate remains poorly understood. In this paper, we introduce VLALagBench, a comprehensive benchmark for studying VLA inference latency sensitivity across task types, object speeds, execution modes (synchronous and asynchronous), and VLA policies. Using VLALagBench, we find that even a modest inference latency of 100 ms can nearly halve task success on moving targets, with sensitivity varying substantially across tasks and policies. These differences reflect task-specific requirements for precision and continuous motion. By overlapping inference with action execution, asynchronous execution can recover some of the lost success at the cost of more frequent inference requests. Building on these findings, we develop an empirical model that predicts task success from limited calibration data, achieving root-mean-square prediction errors of only 5.7-7.8% across various VLA policies. Given an inference system's latency and per-robot throughput, these predictions guide the selection of system-specific VLA algorithm configurations to maximize task success. We believe VLALagBench and these findings provide a solid foundation for future robotics algorithm-system co-design.

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/2611.03103 — 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 2026-11-03T18:17:07.770Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).