TMT: Runtime Backdoor Detection for Vision-Language-Action Policies on Unseen Tasks

Published October 8, 2026 · Category: Robotics

Overview

arXiv:2610.09462v1 Announce Type: new Abstract: Backdoored vision-language-action (VLA) policies can preserve benign task performance while producing malicious actions when a trigger appears. Detecting such activation is difficult because malicious behavior can comprise individually plausible actions, while unfamiliar tasks introduce legitimate changes in observations and behavior. We introduce TMT, a runtime backdoor detector based on Token Manifold and latent Transition modeling. Trained on benign rollouts, its two branches assess input-token structure and prediction errors in adjacent-layer latent dynamics. A suspicious rollout identified by the token manifold branch, once confirmed through latent deviations, guides transition selection for subsequent monitoring. We further explore policy purification through self-distillation: a frozen copy of the backdoored policy provides benign-input actions to supervise a student on paired benign and triggered observations, without requiring a separate clean reference policy. For evaluation, we adapt traditional backdoor detectors and repurpose anomaly and failure detection methods as VLA backdoor detectors. In a post-hoc comparison with ten baselines, TMT achieves state-of-the-art backdoor detection performance on unseen tasks across three VLA backdoor attacks. Our project page is available at https://zzr42.github.io/tmt/.

Source

Originally published at arxiv.org.

Related Articles

• SAFE: Unified Slip and Fracture Detection with Low-Cost Acoustic Sensing in Robotic Grasping
• From Wearable Interfaces to Dexterous Policies: Contact Shifts and Tactile Representations
• HULK: Learning Whole-Body Forceful Loco-Manipulation for Humanoids