Data Party THU

Nov 11, 2025 · Artificial Intelligence

Why Early Adversarial Attacks Still Beat Modern Ones: A Fair Transferability Study

This paper systematically evaluates 23 transferable adversarial attacks and 11 defenses on ImageNet, revealing that early methods like DI outperform many newer attacks when hyper‑parameters are fairly matched, that diffusion‑based defenses give a false sense of security, and that higher transferability often comes at the cost of reduced stealthiness.