Think Twice, Act Once: Token-Aware Compression and Action Reuse for Efficient Inference in Vision-Language-Action Models

Xudong Tan, Yaoxin Yang, Peng Ye, Jialin Zheng, Bizhe Bai, Xinyi Wang, Jia Hao, Tao Chen

Keywords: Vision-Language-Action Models, Model Acceleration

TL;DR: FlashVLA is the first acceleration framework to enable action reuse in Vision-Language-Action models, featuring a training-free and plug-and-play design that significantly reduces inference cost through temporal and visual redundancy exploitation.

Abstract:

Vision-Language-Action (VLA) models have emerged as a powerful paradigm for robot control through natural language instructions. However, their high inference cost—stemming from large-scale token computation and autoregressive decoding—poses significant challenges for real-time deployment and edge applications. While prior work has primarily focused on efficient architectural optimization, we take a different and innovative perspective by identifying a dual form of redundancy in VLA models: (i) high similarity across consecutive action steps, and (ii) substantial redundancy in visual tokens.
Motivated by these observations, we propose FlashVLA, the first training-free and plug-and-play acceleration framework that enables action reuse in VLA models. Specifically, FlashVLA improves inference efficiency through a token-aware action reuse mechanism that avoids redundant decoding across stable action steps, and an information-guided visual token selection strategy that prunes low-contribution tokens.
Extensive experiments on the LIBERO benchmark show that FlashVLA reduces FLOPs by 55.7% and latency by 36.0%, with only a 0.7% drop in task success rate. These results demonstrate the effectiveness of FlashVLA in enabling lightweight, low-latency VLA inference without retraining.

Primary Area: applications to robotics, autonomy, planning

Submission Number: 8280