Near Out-of-Distribution Detection on CIFAR-100 (ID) vs CIFAR-10 (OOD)
0.9623Mahalanobis AUROCR50+ViT-B_16
Evaluation results
| Method | Links | ||
|---|---|---|---|
| R50+ViT-B_16Backbone=R50+ViT-B_16, Pre-trained=ImageNet-21k, In-distribution=CIFAR-100, Out-distribution=CIFAR-102021.06 | 0.9623 | 0.9208 | |
| ViT-B_16Backbone=ViT-B_16, Pre-trained=ImageNet-21k, In-distribution=CIFAR-100, Out-distribution=CIFAR-102021.06 | 0.9553 | 0.9189 | |
| MLP-Mixer-B_16Backbone=MLP-Mixer-B_16, Pre-trained=ImageNet-21k, In-distribution=CIFAR-100, Out-distribution=CIFAR-102021.06 | 0.9531 | 0.9022 | |
| BiT-M R101x3Backbone=BiT-M R101x3, Pre-trained=ImageNet-21k, In-distribution=CIFAR-100, Out-distribution=CIFAR-102021.06 | 0.901 | 0.8369 | |
| BiT-M R50x1Backbone=BiT-M R50x1, Pre-trained=ImageNet-21k, In-distribution=CIFAR-100, Out-distribution=CIFAR-102021.06 | 0.8171 | 0.8115 |