A Unified Evaluation Framework of Physical Adversarial Robustness in the Edge VLM Era

Key points

• However, the physical adversarial robustness of edge VLMs has not been systematically studied, and the fundamental differences in output with modern object detectors create methodological challenges for fair evaluation.1
• To compare the two families directly, we evaluate every selected representative model on one outcome: whether an adversarial patch worn on a person evades detection of the model.1
• We report this by introducing a unified metric, conditional attack success rate (cASR), counted only on images the model still localizes under a matched gray occluder, and partition every failure into omission (no box), substitution (a box on the patch), and displacement (a box on neither).1
• We benchmark the physical adversarial vulnerability of compact grounding VLMs against a range of object detectors, and find that the VLMs can be as vulnerable as the detectors.1

Sources (1)

• [1]A Unified Evaluation Framework of Physical Adversarial Robustness in the Edge VLM Era
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 21, 04:38 AM

  Recent advancements in Vision-Language Models (VLMs) enable end-to-end visual grounding via language prompts and generative decoding, demonstrating largely improved zero-shot and open-vocabulary detection capabilities.

  However, the physical adversarial robustness of edge VLMs has not been systematically studied, and the fundamental differences in output with modern object detectors create methodological challenges for fair evaluation.

Extractive summary: sentences quoted from the sources.