A single object-bound adversarial texture, optimized through a differentiable renderer with an action-space target, steers multitask VLA policies toward attacker-chosen actions and reduces task success across a suite.
OpenVLA: An Open-Source Vision-Language-Action Model , series =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
UniTexture: Cross-Task Universal Adversarial Textures for Vision-Language-Action Models
A single object-bound adversarial texture, optimized through a differentiable renderer with an action-space target, steers multitask VLA policies toward attacker-chosen actions and reduces task success across a suite.