Image Classification on ImageNet (val) (Accuracy and ECE)
79AccuracyEnsemble
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Ensemble2021.05 | 79 | 2.3 | |
| DEnsBackbone=ResNet-502022.07 | 78.1 | — | |
| EnD22021.05 | 77.1 | 1.6 | |
| EnD2021.05 | 77 | 1.6 | |
| VanillaBackbone=ResNet-502022.07 | 76.1 | — | |
| SVIBackbone=ResNet-502022.07 | 76.1 | — | |
| Δ-UQBackbone=ResNet-502022.07 | 76.1 | — | |
| Single2021.05 | 75.9 | 4.8 | |
| MC DropoutBackbone=ResNet-502022.07 | 75 | — | |
| TPTBackbone=CLIP-ViTB/162025.03 | 69 | 10.6 | |
| C-TPTBackbone=CLIP-ViTB/162025.03 | 68.5 | 3.15 | |
| Robust-adapt-SaLs-CTPTBackbone=CLIP-ViTB/162025.03 | 68.04 | 2.63 | |
| Robust-adapt-Penalty-CTPTBackbone=CLIP-ViTB/162025.03 | 68.04 | 2.63 | |
| Robust-adapt-ZS-CTPTBackbone=CLIP-ViTB/162025.03 | 68.01 | 3.01 | |
| O-TPTBackbone=CLIP-ViTB/162025.03 | 67.33 | 1.96 | |
| Zero ShotBackbone=CLIP-ViTB/162025.03 | 66.7 | 2.12 | |
| TPTBackbone=CLIP RN502025.03 | 60.7 | 11.4 | |
| C-TPTBackbone=CLIP RN502025.03 | 60.2 | 3.01 | |
| Robust-adapt-Penalty-CTPTBackbone=CLIP RN502025.03 | 60.06 | 5.93 | |
| Robust-adapt-SaLs-CTPTBackbone=CLIP RN502025.03 | 60.02 | 2.21 | |
| Robust-adapt-ZS-CTPTBackbone=CLIP RN502025.03 | 60 | 2.85 | |
| O-TPTBackbone=CLIP RN502025.03 | 58.97 | 3.1 | |
| Zero ShotBackbone=CLIP RN502025.03 | 58.1 | 2.09 |