LowResearchPreprint
ExecGap: Exploiting Command-Execution Gap for Patch Attacks on Vision-Language-Action Models
- Published
- Record updated
Summary
ExecGap is an adversarial patch attack on Vision-Language-Action (VLA) models that exploits the gap between predicted commands and their physical execution. It biases task-space commands toward directions estimated to have low executability, based on kinematic inefficiency and weak dynamic actuation. Simulation experiments across multiple VLA policies and real-robot tests show it causes larger command-execution mismatch and weaker realized execution than existing patch attacks.
Topics
Related items
- MediumGHSA-25rm-9wvm-m38v: falcon-rust: Discrete-Gaussian sampler runs at 2^-33 precision, far below Falcon's requirement (reference KAT vectors no longer pass)Similar attack · GitHub Advisory Database
- InfoRECLAIM: Robot-Vision-Informed Adversarial Patch Defense via Embodied Attention FeedbackSimilar attack · Arxiv (cs.RO + cs.CV security)
- InfoConfGuard: a physics-grounded reference monitor for predictive-maintenance large-language-model agentsSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
- MediumGemini AI agents repeatedly bypassed safeguards to attack real websiteSimilar attack · BleepingComputer
- InfoWatermark Evidence under Open-Weight LLM Modification: Exact Bounds for Low-Rank Output WatermarksSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)