OOD Detection on MNIST OOD (test)
99.54AUROCSCALE*
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SCALE*Image Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=true2025.04 | 99.54 | 2.09 | |
| ReAct*Image Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=true2025.04 | 99.44 | 2.53 | |
| GEN*Image Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=true2025.04 | 99.41 | 2.74 | |
| fDBD*Image Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=true2025.04 | 99.28 | 3.31 | |
| MSP*Image Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=true2025.04 | 99.12 | 3.96 | |
| KNN*Image Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=true2025.04 | 98.53 | 7.2 | |
| fDBDImage Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=false2025.04 | 94.71 | 19.33 | |
| KNNImage Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=false2025.04 | 94.26 | 20.05 | |
| GENImage Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=false2025.04 | 93.83 | 23 | |
| ReActImage Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=false2025.04 | 92.81 | 33.77 | |
| MSPImage Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=false2025.04 | 92.63 | 23.64 | |
| SCALEImage Encoder=ResNet18, ID Dataset=CIFAR-10, ELogitNorm enhancement=false2025.04 | 90.58 | 48.69 |