# Detect and Suppress: A Mechanistic Defense Against Adversarial Patches on VLA Models

- Source: arXiv cs.RO
- Published: 2026-11-02T05:00:00.000Z
- RoboHot: https://robohot.news/en/items/wengze55y2kdi2kl161tpmvsd
- Original: https://arxiv.org/abs/2610.03498

## Summary

The authors use a sparse autoencoder (SAE) to analyze the internal representations of a VLA model, identify a feature whose activation strongly correlates with the presence of adversarial patches, and suppress that feature at inference time only when a linear probe detects an attack, thereby improving robustness without fine-tuning the VLA. In adversarial patch attack tests on LIBERO-10, conditional intervention improved the success rate under intermittent attacks, whereas continuously applying the same intervention substantially degraded policy performance.