| If your data distribution shifts, use self-learning | ✓ Link | 22.0 | EfficientNet-L2+RPL | 2021-04-27 |
| If your data distribution shifts, use self-learning | ✓ Link | 23.0 | EfficientNet-L2+ENT | 2021-04-27 |
| If your data distribution shifts, use self-learning | ✓ Link | 34.8 | ResNeXt101 32x8d + DeepAug + Augmix + RPL | 2021-04-27 |
| If your data distribution shifts, use self-learning | ✓ Link | 35.5 | ResNeXt101 32x8d + DeepAug + Augmix + ENT | 2021-04-27 |
| Improving robustness against common corruptions by covariate shift adaptation | ✓ Link | 38.0 | ResNeXt101+DeepAug+AugMix, BatchNorm Adaptation, full adaptation | 2020-06-30 |
| Improving robustness against common corruptions by covariate shift adaptation | ✓ Link | 40.7 | ResNeXt101+DeepAug+AugMix, BatchNorm Adaptation, 8 samples | 2020-06-30 |
| If your data distribution shifts, use self-learning | ✓ Link | 40.8 | ResNeXt101 32x8d + IG-3.5B + ENT | 2021-04-27 |
| If your data distribution shifts, use self-learning | ✓ Link | 40.9 | ResNeXt101 32x8d + IG-3.5B + RPL | 2021-04-27 |
| If your data distribution shifts, use self-learning | ✓ Link | 43.2 | ResNeXt101 32x8d + RPL | 2021-04-27 |
| If your data distribution shifts, use self-learning | ✓ Link | 44.3 | ResNeXt101 32x8d + ENT | 2021-04-27 |
| Improving robustness against common corruptions by covariate shift adaptation | ✓ Link | 45.4 | ResNet50+DeepAug+AugMix, BatchNorm Adaptation, full adaptation | 2020-06-30 |
| Improving robustness against common corruptions by covariate shift adaptation | ✓ Link | 48.4 | ResNet50+DeepAug+AugMix, BatchNorm Adaptation, 8 samples | 2020-06-30 |
| If your data distribution shifts, use self-learning | ✓ Link | 50.5 | ResNet50 + RPL | 2021-04-27 |
| If your data distribution shifts, use self-learning | ✓ Link | 51.6 | ResNet50 + ENT | 2021-04-27 |
| Improving robustness against common corruptions by covariate shift adaptation | ✓ Link | 62.2 | ResNet50 (baseline), BatchNorm Adaptation, full adaptation | 2020-06-30 |
| Improving robustness against common corruptions by covariate shift adaptation | ✓ Link | 65.0 | ResNet50 (baseline), BatchNorm Adaptation, 8 samples | 2020-06-30 |