Classification On Imagenet C Ood Class Out Of
Metrics
Detection AUROC (severity 0)
Detection AUROC (severity 10)
Detection AUROC (severity 5)
Results
Performance results of various models on this benchmark
Model Name | Detection AUROC (severity 0) | Detection AUROC (severity 10) | Detection AUROC (severity 5) | Paper Title | Repository |
|---|---|---|---|---|---|
| ViT-L/32-384 with Max-logit | 0.9958 | 0.7748 | 0.9632 | A framework for benchmarking class-out-of-distribution detection and its application to ImageNet | |
| ViT-L/32-384 with Softmax | 0.9915 | 0.7 | 0.9293 | A framework for benchmarking class-out-of-distribution detection and its application to ImageNet | |
| ViT-L/32-384 with MC Dropout | 0.9947 | 0.712 | 0.9478 | A framework for benchmarking class-out-of-distribution detection and its application to ImageNet | |
| ViT-L/32-384 with Entropy | 0.9948 | 0.7332 | 0.9514 | A framework for benchmarking class-out-of-distribution detection and its application to ImageNet | |
| ViT-L/32-384 with ODIN | 0.9955 | 0.7635 | 0.9589 | A framework for benchmarking class-out-of-distribution detection and its application to ImageNet |
0 of 5 row(s) selected.