cs.RONov 18, 2026

RECLAIM: Robot-Vision-Informed Adversarial Patch Defense via Embodied Attention Feedback

Authors: Jiayou Zhu, Peng Guo, Litian Zhang, Xinze Shi, Yuting Zhang, Zelong Yu, Yize Tian, Jirong Liu, 

Organizations: Beijing University of Posts and Telecommunications · Beihang University

Abstract

Vision-language-action policies control embodied agents through a perception-action loop grounded in physical interaction, where adversarial patches can corrupt action predictions and thereby alter subsequent observations. To mitigate these threats, many existing defenses rely on policy adaptation or auxiliary models, but require additional training or inference. We introduce RECLAIM, a training-free, inference-time defense that uses a frozen VLA's own attention response to local visual interventions to guide purification during closed-loop embodied manipulation. Our key insight is that attention changes induced by local edits provide feedback for choosing where and how to purify as the robot interacts with its environment. RECLAIM combines this feedback with robot geometry to refine patch localization and select local treatments, then revalidates them as the agent's embodied actions change the observed scene. We evaluate RECLAIM against patches generated by four adversarial attack methods across multiple manipulation task suites under frozen VLA control. Under adversarial patch attacks, RECLAIM succeeds in at least one rollout on 95% of the evaluated tasks and raises rollout success on LIBERO-Object from 0% without defense to 70.0%. Its mean decision time is only 548.9 ms per control step. These results highlight policy-guided visual purification as an effective approach to defending embodied agents against adversarial patches and recovering closed-loop task execution.

Figures & tables

Explore similar work

CardsList
  1. Structure-Aware Robust Fine-Tuning: Defending Vision-Language-Action Robots Against Physical Attention Hijacking

    Aug 4, 2026Jinquan Zhang, Dongfu Yin, Run Yang +3Fine-TuningLanguage-Conditioned Robot Manipulation

  2. Detect and Suppress: A Mechanistic Defense against Adversarial Patches in VLA Models

    Oct 2, 2026Yukiya Horiba, Koshiro Aoki, Shunsuke Yasuki +2Robustness of VLAsAdversarial Patch Attacks

  3. Beyond Patch Removal: Persistent Adversarial Effects in Vision-Language-Action Policies

    Sep 17, 2026Enhao Wu, Fusen Guo, Yuxin Cao +3Adversarial Patch AttacksAdversarial Attacks on VLAs