Image Classification on CIFAR-100 (test) (Clean and AA Metrics)
68.75Clean AccuracyDAJAT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DAJATBackbone=WRN-34-102022.10 | 68.75 | 31.85 | |
| DAJATBackbone=ResNet-182022.10 | 65.45 | 27.69 | |
| AWPBackbone=WRN-34-102022.10 | 62.73 | 29.59 | |
| HATBackbone=WRN-34-102022.10 | 59.95 | 24.55 | |
| AWPBackbone=ResNet-182022.10 | 59.88 | 25.81 | |
| TRADES + TEBackbone=ResNet-182022.10 | 59.35 | 25.27 | |
| HATBackbone=ResNet-182022.10 | 59.19 | 23.26 | |
| SEATBackbone=ResNet-182022.10 | 56.28 | 27.87 |