← Search

Chengpei Wu

2 accepted papers

2025

Weakly Supervised Contrastive Adversarial Training for Learning Robust Features from Semi-supervised Data

CVPR 2025poster

Existing adversarial training (AT) methods often suffer from incomplete perturbation, meaning that not all non-robust features are perturbed when generating adversarial examples (AEs). This results in residual correlations between non-robust features and labels, leading to suboptimal learning of rob…