cs.LGNov 3, 2026

Say the Same, Act Differently: Diagnosing and Steering Text-Action Divergence in Reasoning VLAs

Authors: Zihao Feng, Qingzhao Zhang, Chunyu Xia, Bo Yu, Z. Morley Mao, Ramesh Govindan

Organizations: University of Southern California · University of Arizona · General Motors

Abstract

Reasoning vision-language-action (VLA) models pair physical control with natural-language rationales that supervisors can read to monitor embodied safety. Prior attacks on VLA reasoning corrupt the rationale itself and leave visible evidence in the text. We show that visual perturbations can instead alter the physical action while keeping the generated rationale word-for-word identical. At the shared text/action interface of Alpamayo for driving and InstructVLA for manipulation, our text-neutral action-subspace (TNAS) diagnostic shows that the locally most action-sensitive direction is nearly orthogonal to the active text-loss gradients, with 1.87% and 5.28% norm overlap. At equal hidden-state perturbation norms across both models, interventions along this text-neutral residual shift actions 4.64x to 47.8x more than text-gradient perturbations, whereas text-gradient perturbations induce 21.2x to 277x more text drift. Steering visual perturbations along this direction reaches the same decoupling from the image. Across 154 obstacle-annotated driving scenes in Alpamayo, TNAS-guided attacks induce a new collision under a word-for-word identical rationale in 52.6% of scenes under full-image perturbations and 35.1% under a localized forward-camera patch, compared with 4.55% for unguided action attacks. Relative to a text-preserving baseline, TNAS guidance raises exact rationale preservation from 83.0% to 89.5% and cuts teacher-forced text drift by 68.4% at a comparable collision rate. In InstructVLA, TNAS guidance cuts text drift by 40.5% and preserves the exact response on 85.0% of examples alongside a 1.73-unit action-chunk shift. An unchanged rationale therefore cannot certify a safe action, so monitors for reasoning VLAs must verify the predicted action channel directly.

Figures & tables

Appendix figures & tables7 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. ReasonBreak: Probing Vulnerabilities in Reasoning-Enabled Vision-Language-Action Models for Autonomous Driving

    May 27, 2026Mohammadreza Teymoorianfard, Jean-Philippe Monteuuis, Jonathan Petit +1VLMs for Autonomous DrivingAutonomous Driving Safety Evaluation

  2. TacitVLA: Reasoning Vision-Language-Action Models Without Reasoning Latency

    Oct 13, 2026Jack Lu, Yurong You, Boyi Li +8Reasoning for VLA ModelsVLMs for Robotics