Metric: Top 1 Error (lower is better)
| # | Model↕ | Top 1 Error▲ | Augmentations | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | Model soups (ViT-G/14) | 4.54 | Yes | Model soups: averaging weights of multiple fine-... | 2022-03-10 | Code |
| 2 | EfficientNet-L2 Noisy Student + RPL | 17.4 | Yes | If your data distribution shifts, use self-learn... | 2021-04-27 | Code |
| 3 | EfficientNet-L2 Noisy Student + ENT | 19.7 | Yes | If your data distribution shifts, use self-learn... | 2021-04-27 | Code |
| 4 | ResNeXt101+DeepAug+AugMix, BatchNorm Adaptation, | 44 | No | Improving robustness against common corruptions ... | 2020-06-30 | Code |
| 5 | ResNet50+DeepAug+Augmix, BatchNorm adaptation | 48.9 | No | Improving robustness against common corruptions ... | 2020-06-30 | Code |
| 6 | ResNet50 + RPL | 54.1 | No | If your data distribution shifts, use self-learn... | 2021-04-27 | Code |
| 7 | ResNet50 + ENT | 56.1 | No | If your data distribution shifts, use self-learn... | 2021-04-27 | Code |
| 8 | ResNet50, BatchNorm adaptation | 59.9 | No | Improving robustness against common corruptions ... | 2020-06-30 | Code |