Metraipolis › archive › dc_2c685fc0d294

Toward Real-Time VLAs: Stage-Aware Two-Step Flow Denoising and System-Level Evaluation — arXiv:2609.39822

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.39822
License (as archived): CC0-1.0
Archived: 2026-10-07T15:12:13.497Z
SHA-256 of the archived text: c59ad42a9df41ad55fd63caf29939e6dce07511a92fda095b570dd4f3fc5ac3f
Stable URL of this page: https://metraipolis.com/documento/dc_2c685fc0d294

How to cite

Toward Real-Time VLAs: Stage-Aware Two-Step Flow Denoising and System-Level Evaluation — arXiv:2609.39822 — arXiv (en), https://arxiv.org/abs/2609.39822 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2c685fc0d294 (Metraipolis archive dc_2c685fc0d294, sha256 c59ad42a9df41ad55fd63caf29939e6dce07511a92fda095b570dd4f3fc5ac3f)

BibTeX

@misc{dc2c685fc0d294,
  title = {Toward Real-Time VLAs: Stage-Aware Two-Step Flow Denoising and System-Level Evaluation — arXiv:2609.39822},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.39822},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2c685fc0d294; sha256 c59ad42a9df41ad55fd63caf29939e6dce07511a92fda095b570dd4f3fc5ac3f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39822. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-language-action (VLA) models face a timing gap between low-rate inference and high-rate robot execution. We characterize this gap through end-to-end latency measurements of model inference and the robot execution chain. Repeated Flow Matching denoising contributes substantially to inference cost, while robot-side delays mainly arise from perception acquisition, communication scheduling, and physical response. Analysis of the velocity field shows relatively stable magnitude and direction in early integration, followed by stronger directional correction near the terminal steps. Based on this stage heterogeneity, we propose two-stage non-uniform denoising, reducing the number of steps from 10 to 2 and model-inference time from 61.557 ms to 21.956 ms. We also develop a distributed real-time VLA framework with independent inference, action-publication, and robot-control rates, modular observation acquisition, and action-provenance logging. Using {\pi}0.5 as the baseline, we evaluate six real-time execution methods on a long-horizon physical garment-folding task. Legato performs best overall among training-based methods, while Temporal Smoothing leads among training-free methods; both perform strongly in task success, completion time, action continuity, and acceleration smoothness. Combining two-step denoising with representative execution methods substantially reduces inference cost with a small reduction in task performance. These results motivate joint optimization of model-inference efficiency and robot-system timing.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.39822 — 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 2026-10-07T14:44:47.473Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).