OOD Detection on ImageNet-ES (test)
81.7AUROCMM++
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MM++Backbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 81.7 | 52.43 | |
| X-Maha†Backbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 76.78 | 63.45 | |
| Maha++Backbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 73.51 | 73.09 | |
| rMaha++Backbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 71.9 | 72.93 | |
| rMahaBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 71.3 | 80.22 | |
| ReActBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 69.57 | 67.84 | |
| MahaBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 69.53 | 84.53 | |
| MSPBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 69.35 | 70.01 | |
| EnergyBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 67.54 | 70.31 | |
| ODINBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 63.22 | 74.31 | |
| KNNBackbone=ViT-B/16, ID Dataset=ImageNet-LT2026.06 | 55.88 | 91.43 |