cs.CVNov 17, 2026

VSpark: Speculative Decoding for Autoregressive Vision-Language-Action Models

Authors: Suho Yoo, Youngjoon Jang, Hyebin Cho, Taeyeop Lee, Jaewook Jeong, Joon Son Chung

Organizations: KAIST · VGG, University of Oxford

Abstract

Our aim is to accelerate autoregressive vision-language-action models (VLAs), which generate robot actions one token at a time. Speculative decoding makes this possible: a lightweight draft model proposes several tokens at once, and the VLA verifies them in a single pass, accepting only those that match its own predictions, so decoding remains lossless. The speedup, however, is limited in robot control, where many different tokens encode nearly the same motion and a draft close to the correct action is still rejected. To address this, our approach, VSpark, is built upon three design choices: (i) rollouts collected with calibrated action noise, which expose the draft to a wider range of robot states; (ii) feature alignment, which trains the draft to match the VLA's hidden features rather than its output tokens alone; and (iii) continuation reuse, which keeps the predictions the VLA has already computed along rejected drafts as candidates for the next verification step, instead of discarding them. On four VLAs with different action tokenizers and four manipulation benchmarks, VSpark achieves 1.28x to 2.10x speedup while keeping the VLA's decoding lossless.

Figures & tables

Appendix figures & tables18 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. TokenPB: Enabling VLA Action Self-Verification via VLM Token Probability Bucketing

    Mar 18, 2026Chen Zhao, Zhuoran Wang, Haoyang Li +6Diffusion-Based VLAsVision-Language-Action Models

  2. Reducing Temporal Redundancy for Efficient Vision-Language-Action Inference

    Jul 14, 2026Yuzhou Wu, Yuxin Zheng, Muchun Niu +6Efficient VLA InferenceEfficient VLAs