OOD Detection on ImageNet-1K OOD Benchmarks Overall Average
40.96FPR@95EPD
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| EPDBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 40.96 | 87.26 | |
| RMDSBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 43.4 | 87.6 | |
| MDSBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 44.43 | 87.18 | |
| SHEBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 44.63 | 85.89 | |
| ViMBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 47.02 | 86.51 | |
| KNNBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 47.35 | 84.13 | |
| GENBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 47.65 | 85.33 | |
| MSPBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 63.75 | 81.03 | |
| DICEBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 65.23 | 75.53 | |
| TempScaleBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 66.11 | 81.06 | |
| ReActBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 66.15 | 79.12 | |
| OpenMaxBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 68.81 | 83.02 | |
| MLSBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 84.44 | 77.45 | |
| EBOBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 88.49 | 72.35 | |
| GradNormBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 93.47 | 46.52 | |
| ASHBackbone=ViT-B/16, ID Accuracy=81.14%, ID Dataset=ImageNet-1K2026.04 | 95.84 | 52.22 |