cs.AINov 9, 2026

APS-VLA: Action-Guided and Future-Aware Token Pruning for Efficient Vision-Language-Action Models

Authors: Xiaoyang Xing, Chun-Mei Feng, Anthony Cohn, Xiaoliang Gong, Asoke Nandi

Organizations: School of Computer Science and Technology, Tongji University · Agency for Science, Technology and Research (A*STAR), Singapore · School of Computer Science, University of Leeds · Electronic and Electrical Engineering, Brunel University of London

Abstract

Vision-language-action (VLA) policies have demonstrated strong generalization in robotic manipulation, yet processing dense visual tokens often incurs high inference latency. Visual token pruning offers a practical way to reduce this cost, but identifying a compact set of tokens that preserves reliable action generation remains challenging. Existing methods largely rely on static or immediate heuristic cues to estimate token importance. We introduce APS-VLA, a framework that learns visual token selection from robot demonstrations while keeping the pretrained VLA backbone frozen. A lightweight Decision Influence Estimator scores tokens conditioned on current visual observations, language instructions, and robot proprioception. Its training combines two complementary signals: action supervision matches both the full-input policy predictions and demonstrated actions, while an auxiliary visual transition objective uses demonstrated actions to predict subsequent visual features from the selected representation. At deployment, the estimator performs a single hard selection and packing step before the policy's native action-generation procedure; the auxiliary predictor is removed. Experiments with OpenVLA-OFT on LIBERO and with π0.5π_{0.5} and XR-1 on RoboCasa365 demonstrate inference acceleration with modest changes in task success. Across five real-world tasks on two dual-arm platforms, APS-VLA reduces mean policy inference latency from 137.25 to 93.25 ms, achieving a 1.47×1.47\times speedup with an average success rate of 61%, compared with 60% for the unpruned baseline. Together, these results support APS-VLA as a modular, plug-in framework for efficient robotic control across VLA architectures, preserving the underlying policy parameters and native action-generation procedure.

Figures & tables

Explore similar work

CardsList
  1. ActPrune: Action-Grounded Visual Token Pruning for Efficient Vision-Language-Action Models

    Nov 6, 2026Sungwon Moon, Minji Kim, Hyemin Boo +2Efficient VLA Model InferenceVisual Token Pruning

  2. FastVLA: Task-Conditioned and Temporally Adaptive Visual Token Compression for Efficient Vision-Language-Action Models

    Oct 15, 2026Jiuyi Xu, Chengyuan Qian, Zhengzhong Tu +2Efficient VLA Model InferenceVisual Token Compression