OOD Detection on ImageNet-C (test)
84.35AUROCMM++
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MM++Backbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 84.35 | 53.8 | |
| X-Maha†Backbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 83.96 | 53.57 | |
| Maha++Backbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 74.86 | 73.05 | |
| ReActBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 73.55 | 65.85 | |
| rMaha++Backbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 72.49 | 74.97 | |
| EnergyBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 72.34 | 66.65 | |
| rMahaBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 71.73 | 80.32 | |
| MSPBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 71.21 | 73.94 | |
| MahaBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 70.94 | 85 | |
| ODINBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 64.61 | 79.35 | |
| KNNBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 57.73 | 88.45 |