Distilling Vision-Language Models for Robust Traffic Sign Perception in Autonomous Vehicles

Abstract

Traffic sign recognition (TSR) models based on deep neural networks achieve strong clean-data performance but remain vulnerable to physically realizable adversarial attacks, including shadow perturbations, natural-light interference, and printed patches. Existing defenses often improve robustness against one attack type while degrading performance on others, and can reduce clean accuracy. We propose LAMDA (Language-Anchored Model for Direction Alignment), a training framework that transfers lang

Categories

Framework mappings

Suggested from the entry's categories.

@misc{mohajeransari2026distilling,
  title = {{Distilling Vision-Language Models for Robust Traffic Sign Perception in Autonomous Vehicles}},
  author = {Pedram MohajerAnsari and Amir Salarpour and Mert D. Pesé},
  year = {2026},
  month = aug,
  eprint = {2608.08815},
  archivePrefix = {arXiv},
  url = {https://www.semanticscholar.org/paper/2b667cc2121abfa5bb11c1d1af9c50046c5bc69f}
}