OOD detection on Oxford-IIIT Pet
99.99AUROCANTS
Evaluation results
| Method | Links | ||
|---|---|---|---|
| ANTSZero-shot=true2025.09 | 99.99 | 0.02 | |
| NegLabelZero-shot=true2025.09 | 99.62 | 1.7 | |
| CLS-MBackbone=ViT-B/16, Evaluation Protocol=16-shot Prompt Learning2024.05 | 94.7 | — | |
| CLS-EBackbone=ViT-B/16, Evaluation Protocol=16-shot Prompt Learning2024.05 | 94.58 | — | |
| RMDSBackbone=ViT-B/16, Evaluation Protocol=Fully Fine-tuned2024.05 | 93.26 | — | |
| GradNormBackbone=ViT-B/16, Evaluation Protocol=Fully Fine-tuned2024.05 | 92 | — | |
| ReactBackbone=ViT-B/16, Evaluation Protocol=Fully Fine-tuned2024.05 | 91.73 | — | |
| MaxLogitBackbone=ViT-B/16, Evaluation Protocol=16-shot Prompt Learning2024.05 | 91.64 | — | |
| ASHBackbone=ViT-B/16, Evaluation Protocol=Fully Fine-tuned2024.05 | 91.51 | — | |
| EnergyBackbone=ViT-B/16, Evaluation Protocol=16-shot Prompt Learning2024.05 | 91.23 | — | |
| ODINBackbone=ViT-B/16, Evaluation Protocol=Fully Fine-tuned2024.05 | 91.13 | — | |
| NegPromptBackbone=ViT-B/16, Evaluation Protocol=16-shot Prompt Learning2024.05 | 90.26 | — | |
| NegLabelBackbone=ViT-B/16, Evaluation Protocol=Zero-shot CLIP2024.05 | 89.74 | — | |
| MSPBackbone=ViT-B/16, Evaluation Protocol=Fully Fine-tuned2024.05 | 89.16 | — | |
| AdaNegBackbone=ViT-B/16, Evaluation Protocol=Zero-shot CLIP2024.05 | 88.26 | — | |
| VIMBackbone=ViT-B/16, Evaluation Protocol=Fully Fine-tuned2024.05 | 87.99 | — | |
| LoCoOp (GL-MCM)Backbone=ViT-B/16, Evaluation Protocol=16-shot Prompt Learning2024.05 | 86.62 | — | |
| MDSBackbone=ViT-B/16, Evaluation Protocol=Fully Fine-tuned2024.05 | 86.14 | — | |
| CLIPNBackbone=ViT-B/16, Evaluation Protocol=Zero-shot CLIP2024.05 | 82.65 | — | |
| MCMBackbone=ViT-B/16, Evaluation Protocol=Zero-shot CLIP2024.05 | 77.29 | — | |
| KNNBackbone=ViT-B/16, Evaluation Protocol=Fully Fine-tuned2024.05 | 73.93 | — |