TAKO demonstrates real-time adversarial takeover of robotic diffusion policies via reusable universal patches on visual inputs, achieving 100% success in steering attacker-chosen trajectories across multiple tasks, encoders, and diffusion methods.
Exploring the adversarial vulnerabilities of vision- language-action models in robotics, 2025a
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
representative citing papers
A prompt-only attack called command-preserving trajectory redirection can steer VLA robot behavior to attacker-chosen physical outcomes while the text still appears to match the intended task.
ROBOSHACKLES is a new safety dataset for embodied foundation models created through a pipeline of hazard-aware editing and video synthesis from real observations, with all six tested models generating unsafe actions at a 100% rate.
VLA-Hijack is a new adversarial patch attack on Vision-Language-Action models that suppresses real arm features and injects the patch as surrogate embodiment to achieve high cross-architecture transferability.
SOTA VLA models like OpenVLA and pi0 collapse when object positions are perturbed, indicating that standard LIBERO scores reward memorization; LIBERO-PRO provides a systematic perturbed evaluation suite.
citing papers explorer
-
Test-time Adversarial Takeover: A Real-time Hijacking Interface against Robotic Diffusion Policies
TAKO demonstrates real-time adversarial takeover of robotic diffusion policies via reusable universal patches on visual inputs, achieving 100% success in steering attacker-chosen trajectories across multiple tasks, encoders, and diffusion methods.
-
Trajectory-Level Redirection Attacks on Vision-Language-Action Models
A prompt-only attack called command-preserving trajectory redirection can steer VLA robot behavior to attacker-chosen physical outcomes while the text still appears to match the intended task.
-
ROBOSHACKLES: A Safety Dataset for Human-Injury Prevention in Embodied Foundation Models
ROBOSHACKLES is a new safety dataset for embodied foundation models created through a pipeline of hazard-aware editing and video synthesis from real observations, with all six tested models generating unsafe actions at a 100% rate.
-
VLA-Hijack: A Transferable Patch Attack against Vision-Language-Action Models via Visual Proprioception Hijacking
VLA-Hijack is a new adversarial patch attack on Vision-Language-Action models that suppresses real arm features and injects the patch as surrogate embodiment to achieve high cross-architecture transferability.
-
LIBERO-PRO: Towards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorization
SOTA VLA models like OpenVLA and pi0 collapse when object positions are perturbed, indicating that standard LIBERO scores reward memorization; LIBERO-PRO provides a systematic perturbed evaluation suite.