AI & ML
ExecGap: Exploiting Command-Execution Gap for Patch Attacks on Vision-Language-Action Models
arXiv cs.RO0 views
arXiv:2611.23714v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have become a powerful paradigm for robotic manipulation and show strong promise for real-world deployment, making adversarial robustness critical for reliable robot operation. Existing adversarial attacks on VLA-controlled robots mainly optimize prediction-level discrepancies or rollout-level deviations to disrupt manipulation. However, they rarely account for the fact that VLA-predicted commands are not executed directly, but must be realized through downstream control and robot embodiment. In this paper, we propose ExecGap, an adversarial patch attack framework that exploits this command-execution gap to expose embodiment-specific execution vulnerabilities. Specifically, ExecGap biases VLA-predicted task-space commands toward low-executability directions, which are estimated from kinematic inefficiency and dynamic weak actuation. This induces execution degradation, where the predicted command becomes difficult to realize effectively, thereby impairing task completion. Extensive simulation experiments across multiple VLA policies show that ExecGap induces larger command-execution mismatch and weaker realized execution than existing patch attacks. Real-robot experiments further validate that the deployed patch can induce execution degradation on physical platforms.
Source:https://arxiv.org/abs/2611.23714PreviousMeasuring the Conservatism of Latent Safety Filters for World ModelsNextImproving While Deployed: Learning Proactive End-to-End Driving from Imperfect Human Takeover Data
Related posts

Guardrails and cost control for your agentic workloads with Amazon Redshift
0AWS Big Data Blog

Fast failover with Amazon Route 53 Private Endpoint Health Checks
0AWS Networking Blog

PolyAI Dialog-RSN-1 is now on Amazon SageMaker JumpStart
0AWS Machine Learning Blog

Keep inference running through regional capacity shortfalls with cross-Region inference routing in Amazon SageMaker AI
0AWS Machine Learning Blog

From learning to leaderboard: How Cisco upskilled 1,200 engineers in agentic AI with AWS AI League
0AWS Machine Learning Blog

Optimize AI economics on Amazon Bedrock with Accenture’s sequenced approach
0AWS Machine Learning Blog