Recent Advances in Vision Transformer Robustness Against Adversarial Attacks in Traffic Sign Detection and Recognition: A Survey
Oluwajuwon A. Fawole, Danda B. Rawat · ACM Computing Surveys · 2025
The emergence of Vision Transformers (ViTs) has marked a significant advancement in machine learning, particularly in applications requiring robust visual recognition capabilities, such as traffic sign detection for autonomous driving systems. But, deploying these models in adversarial environments where robustness is critical remains a challenge. This survey provides a comprehensive review of the integration of ViTs in traffic sign detection and recognition, emphasizing their vulnerability to adversarial attacks and the methods developed to enhance their robustness. This article also presents a compressive comparison of ViTs in a tabular form for side-by-side comparison.