mlsanityData to 7 Oct 2026

Adversarial Vertex Mixup: Toward Better Adversarially Robust Generalization

arXiv 2003.02484 ↗First posted 5 Mar 2020cs.CVcited by 73

Not on the map. Why?

Abstract (arXiv)

Adversarial examples cause neural networks to produce incorrect outputs with high confidence. Although adversarial training is one of the most effective forms of defense against adversarial examples, unfortunately, a large gap exists between test accuracy and training accuracy in adversarial training. In this paper, we identify Adversarial Feature Overfitting (AFO), which may cause poor adversarially robust generalization, and we show that adversarial training can overshoot the optimal point in terms of robust generalization, leading to AFO in our simple Gaussian model. Considering these theoretical results, we present soft labeling as a solution to the AFO problem. Furthermore, we propose Adversarial Vertex mixup (AVmixup), a soft-labeled data augmentation approach for improving adversarially robust generalization. We complement our theoretical analysis with experiments on CIFAR10, CIFAR100, SVHN, and Tiny ImageNet, and show that AVmixup significantly improves the robust generalization performance and that it reduces the trade-off between standard accuracy and adversarial robustness.

Key facts. Cited by 73 papers in the map (data through 7 Oct 2026). First posted on arXiv 5 Mar 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 22; 2022: 19; 2023: 18; 2024: 7; 2025: 6; 2026: 1.

73 papers.

Sort:

  1. 2021Recent Advances in Adversarial Training for Adversarial Robustnesscited by 276
  2. 2023Cross-Entropy Loss Functions: Theoretical Analysis and Applicationscited by 206
  3. 2021Admix: Enhancing the Transferability of Adversarial Attackscited by 155
  4. 2021Advances in adversarial attacks and defenses in computer vision: A surveycited by 127
  5. 2022LAS-AT: Adversarial Training with Learnable Attack Strategycited by 94
  6. 2022A Light Recipe to Train Robust Vision Transformerscited by 64
  7. 2021Adversarial Robustness under Long-Tailed Distributioncited by 50
  8. 2022Understanding Robust Overfitting of Adversarial Training and Beyondcited by 50
  9. 2021SmoothMix: Training Confidence-calibrated Smoothed Classifiers for Certified Robustnesscited by 49
  10. 2021Adversarial Attacks On Multi-Agent Communicationcited by 48
  11. 2022Enhancing Adversarial Training with Second-Order Statistics of Weightscited by 42
  12. 2022Sparsity Winning Twice: Better Robust Generalization from More Efficient Trainingcited by 41
38 more
  1. 2021Boosting Fast Adversarial Training with Learnable Adversarial Initializationcited by 40
  2. 2023Improving Fast Adversarial Training with Prior-Guided Knowledgecited by 34
  3. 2022A Survey of Mix-based Data Augmentation: Taxonomy, Methods, Applications, and Explainabilitycited by 30
  4. 2021Low Curvature Activations Reduce Overfitting in Adversarial Trainingcited by 29
  5. 2021AdvRush: Searching for Adversarially Robust Neural Architecturescited by 28
  6. 2023Randomized Adversarial Training via Taylor Expansioncited by 28
  7. 2022Investigating Top-k White-Box and Transferable Black-box Attackcited by 26
  8. 2021Survey: Image Mixing and Deleting for Data Augmentationcited by 25
  9. 2022Robust Weight Perturbation for Adversarial Trainingcited by 25
  10. 2021Removing Undesirable Feature Contributions Using Out-of-Distribution Datacited by 23
  11. 2023Delving into the Adversarial Robustness of Federated Learningcited by 19
  12. 2024A Survey on Mixup Augmentations and Beyondcited by 18
  13. 2022Robust Trajectory Prediction against Adversarial Attackscited by 16
  14. 2023Generalist: Decoupling Natural and Robust Generalizationcited by 14
  15. 2023Defenses in Adversarial Machine Learning: A Surveycited by 14
  16. 2021LTD: Low Temperature Distillation for Gradient Masking-free Adversarial Trainingcited by 13
  17. 2023Adversarial Training Should Be Cast as a Non-Zero-Sum Gamecited by 13
  18. 2024Revisiting Adversarial Training under Long-Tailed Distributionscited by 13
  19. 2022Why adversarial training can hurt robust accuracycited by 12
  20. 2023Certified Robust Neural Networks: Generalization and Corruption Resistancecited by 11
  21. 2024DAFA: Distance-Aware Fair Adversarial Trainingcited by 11
  22. 2025Enhancing Robust Fairness via Confusional Spectral Regularizationcited by 11
  23. 2021AID-Purifier: A Light Auxiliary Network for Boosting Adversarial Defensecited by 10
  24. 2021Interpolated Joint Space Adversarial Training for Robust and Generalizable Defensescited by 10
  25. 2021MixACM: Mixup-Based Robustness Transfer via Distillation of Activated Channel Mapscited by 9
  26. 2023It Is All About Data: A Survey on the Effects of Data on Adversarial Robustnesscited by 9
  27. 2021NoiLIn: Improving Adversarial Training and Correcting Stereotype of Noisy Labelscited by 8
  28. 2022A.I. Robustness: a Human-Centered Perspective on Technological Challenges and Opportunitiescited by 8
  29. 2023Fast Propagation is Better: Accelerating Single-Step Adversarial Training via Sampling Subnetworkscited by 8
  30. 2021Adversarially Optimized Mixup for Robust Classificationcited by 7
  31. 2021Domain Invariant Adversarial Learningcited by 7
  32. 2024Efficient local linearity regularization to overcome catastrophic overfittingcited by 7
  33. 2022MixupE: Understanding and Improving Mixup from Directional Derivative Perspectivecited by 6
  34. 2024Conserve-Update-Revise to Cure Generalization and Robustness Trade-off in Adversarial Trainingcited by 6
  35. 2021Regional Adversarial Training for Better Robust Generalizationcited by 4
  36. 2023Revisiting the Trade-off between Accuracy and Robustness via Weight Distribution of Filterscited by 4
  37. 2023Splitting the Difference on Adversarial Trainingcited by 4
  38. 2021Adversarial Mixing Policy for Relaxing Locally Linear Constraints in Mixupcited by 3

50 most cited of 73 All 73 (JSON)