cs.RONov 17, 2026

SR-VLA: Stall-Reduced VLA Execution via Predicted-Input Lookahead and Selective KV Cache Repair

Authors: Tao Zhang, Yi Zhong, Jinhao Ruan, Yipeng Gao, Linfeng Zhong, Youming Yang, Yuan Wang

Organizations: School of Integrated Circuits, Peking University · Key Laboratory of Microelectronic Devices and Circuits, MPW Center, Peking University · Beijing Advanced Innovation Center for Integrated Circuits

Abstract

Vision-language-action (VLA) policies incur control stalls when inference blocks the next action chunk. Asynchronous early inference can hide latency, but its results may be inconsistent with the arriving actual inputs. Unconditional reuse risks acting on mismatched inputs, whereas full recomputation discards useful precomputation. We propose SR-VLA, combining predicted-input lookahead with selective KV cache repair. During current chunk execution, a lightweight action-conditioned predictor generates next-boundary visual observations and proprioceptive state. These predictions, together with the task instruction, drive the frozen VLA to precompute a speculative KV cache and action chunk. After real inputs are available to VLA, predicted--real differences and task conditions determine whether to accept the action chunk or trigger selective KV repair. On LIBERO, SR-VLA achieves 2.67×2.67\times and 1.27×1.27\times query speedups over the respective synchronous π0.5π_{0.5} and OpenVLA-OFT baselines, while maintaining competitive average task success rates. Under target-motion perturbation, success remains baseline-comparable, with 1.52×1.52\times--1.93×1.93\times query speedup.

Figures & tables

Explore similar work

CardsList
  1. Neural Introspection Gating for Adaptive KV-Cache Reuse in Vision-Language-Action Models

    Aug 11, 2026Zhijie Wu, Kento Kawaharazuka, Kei OkadaAdaptive InferenceKV Caching

  2. Leap-VLA: High-Rate Robot Control from a Low-Rate VLM Backbone

    Oct 31, 2026Lin Yang, Yunjie Shen, Zheyu Li +8Closed-Loop Control of VLA ModelsEfficient VLA Model Inference

  3. What Makes a VLA Reactive? A Study of Execution Strategies for Dynamic Manipulation

    Nov 11, 2026Zhuoyue Li, Zhendong Han, Pankhuri Vanjani +2Vision-Language-Action ModelsClosed-Loop Control of VLA Models