Hong-Bin Zhang

CaptchaBench: A Modality-Stratified Benchmark Dataset for Evaluating Adversarial Perturbation Against VLM-based CAPTCHA Recognition

Vision-Language Models (VLMs) now achieve over 99% zero-shot accuracy on Chinese-character CAPTCHAs, rendering traditional hardening strategies (complex backgrounds, stylized fonts, character distortion) ineffective. Adversarial perturbation offers an emerging defense by exploiting model vulnerabilities at the decision...