Papers › Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria
Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria
Nuoyan Zhou, Nannan Wang, Decheng Liu, Dawei Zhou, Xinbo Gao
Deep neural networks are vulnerable to adversarial noise. Adversarial Training (AT) has been demonstrated to be the most effective defense strategy to protect neural networks from being fooled. However, we find AT omits to learning robust features, resulting in poor performance of adversarial robustness. To address this issue, we highlight two criteria of robust representation: (1) Exclusion: \emph{the feature of examples keeps away from that of other classes}; (2) Alignment: \emph{the feature of natural and corresponding adversarial examples is close to each other}. These motivate us to propose a generic framework of AT to gain robust representation, by the asymmetric negative contrast and reverse attention. Specifically, we design an asymmetric negative contrast based on predicted probabilities, to push away examples of different classes in the feature space. Moreover, we propose to weight feature by parameters of the linear classifier as the reverse attention, to obtain class-aware feature and pull close the feature of the same class. Empirical evaluations on three benchmark datasets show our methods greatly advance the robustness of AT and achieve state-of-the-art performance.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
2 archive task tags without a task page not shown.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Adversarial Attack | CIFAR-10 | TRADES-ANCRA/ResNet18 | Attack: AutoAttack | 59.70 | #5 of 6 | Archive leaderboard | report |
| Adversarial Defense | CIFAR-10 | ResNet18 (TRADES-ANCRA/PGD-40) | Accuracy | 81.70 | #6 of 8 | Archive leaderboard | report |
| Adversarial Defense | CIFAR-10 | ResNet18 (TRADES-ANCRA/PGD-40) | Attack: AutoAttack | 59.70 | #6 of 8 | Archive leaderboard | report |
| Adversarial Defense | CIFAR-10 | ResNet18 (TRADES-ANCRA/PGD-40) | Robust Accuracy | 82.96 | #6 of 8 | Archive leaderboard | report |
| Adversarial Defense | CIFAR-100 | resnet18 | autoattack | 60.10/35.05 | #3 of 3 | Archive leaderboard | report |
| Adversarial Robustness | CIFAR-10 | TRADES-ANCRA/ResNet18 | Accuracy | 81.70 | #5 of 5 | Archive leaderboard | report |
| Adversarial Robustness | CIFAR-10 | TRADES-ANCRA/ResNet18 | Attack: AutoAttack | 59.70 | #5 of 5 | Archive leaderboard | report |
| Adversarial Robustness | CIFAR-100 | ResNet18/MART-ANCRA | AutoAttacked Accuracy | 35.05 | #2 of 2 | Archive leaderboard | report |
| Adversarial Robustness | CIFAR-100 | ResNet18/MART-ANCRA | Clean Accuracy | 60.10 | #2 of 2 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections