 Sparse-PGD: A Unified Framework for Sparse Adversarial Perturbations GenerationIEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2024 |
 On the Over-Memorization During Natural, Robust and Catastrophic
OverfittingInternational Conference on Learning Representations (ICLR), 2023 |
 Fast Adversarial Training with Smooth ConvergenceIEEE International Conference on Computer Vision (ICCV), 2023 |
 Robustness and Accuracy Could Be Reconcilable by (Proper) DefinitionInternational Conference on Machine Learning (ICML), 2022 |
 Make Some Noise: Reliable and Efficient Single-Step Adversarial TrainingNeural Information Processing Systems (NeurIPS), 2022 |
