OOD Detection on CIFAR-100 SCOOD Benchmark Average (6 OOD datasets)
41.05FPR@95%ET-OOD
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| ET-OODBackbone=ResNet-18, Extra Unlabeled Training Data=Tiny-ImageNet2023.03 | 41.05 | 82.44 | 84.37 | 76.47 | 11.7 | 19.47 | 34.05 | 50.97 | |
| UDGBackbone=ResNet-18, Extra Unlabeled Training Data=Tiny-ImageNet2023.03 | 75.45 | 79.63 | 80.69 | 74.1 | 3.85 | 8.66 | 20.57 | 44.47 | |
| OEBackbone=ResNet-18, Extra Unlabeled Training Data=Tiny-ImageNet2023.03 | 80.06 | 78.46 | 80.22 | 71.83 | 2.74 | 8.37 | 22.18 | 46.75 | |
| EBOBackbone=ResNet-18, Extra Unlabeled Training Data=None2023.03 | 81.66 | 79.31 | 80.54 | 72.82 | 2.43 | 7.26 | 21.41 | 49.39 | |
| ODINBackbone=ResNet-18, Extra Unlabeled Training Data=None2023.03 | 81.89 | 77.98 | 78.54 | 72.56 | 1.84 | 5.65 | 17.77 | 46.73 | |
| MCDBackbone=ResNet-18, Extra Unlabeled Training Data=Tiny-ImageNet2023.03 | 85.14 | 74.82 | 75.93 | 69.14 | 1.06 | 4.6 | 16.73 | 41.83 |