Out Of Distribution Detection On Imagenet 1K 3
평가 지표
AUROC
FPR95
Latency, ms
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | AUROC | FPR95 | Latency, ms |
|---|---|---|---|
| nearest-neighbor-guidance-for-out-of-1 | 97.7 | 11.12 | 11.10 |
| nearest-neighbor-guidance-for-out-of-1 | 99.57 | 1.83 | 31.00 |
| delving-into-out-of-distribution-detection | 94.95 | 28.38 | - |
| on-the-importance-of-gradients-for-detecting | - | 50.03 | - |
| neuron-activation-coverage-rethinking-out-of | 96.52 | - | - |
| on-the-effectiveness-of-sparsification-for | 96.24 | 18.64 | - |
| line-out-of-distribution-detection-by | 97.56 | 12.26 | - |
| out-of-distribution-detection-with-deep | 94.72 | 30.83 | - |
| out-of-distribution-detection-with-deep | 86.2 | 59.08 | - |
| rethinking-out-of-distribution-ood-detection | 86.9 | - | - |
| detecting-out-of-distribution-data-through-in | - | 43.87 | - |
| react-out-of-distribution-detection-with-1 | 91.53 | 42.40 | - |
| mos-towards-scaling-out-of-distribution | 98.15 | 9.28 | - |
| gen-pushing-the-limits-of-softmax-based-out | - | 68.32 | - |
| extremely-simple-activation-shaping-for-out | 97.87 | 11.49 | - |
| non-parametric-outlier-synthesis | 96.19 | 16.58 | - |
| delving-into-out-of-distribution-detection | 94.61 | 30.91 | - |
| boosting-out-of-distribution-detection-with | 97.67 | - | - |
| forte-finding-outliers-with-representation | 99.67 ± 00.03 | 0.64 ± 00.06 | - |
| unleashing-mask-explore-the-intrinsic-out-of | 94.71 | 21.97 | - |
| rankfeat-rank-1-feature-removal-for-out-of | 91.91 | 41.31 | - |
| decoupling-maxlogit-for-out-of-distribution | - | 47.32 | - |
| out-of-distribution-detection-based-on-in | - | 34.22 | - |
| vim-out-of-distribution-with-virtual-logit | 99.41 | 2.60 | - |
| discopatch-batch-statistics-are-all-you-need | 99.1 | 3.6 | 1.56 |
| scaling-for-training-time-and-post-hoc-out-of | 98.17 | 9.5 | 11.27 |
| out-of-distribution-detection-with-implicit | 85.98 | 55.87 | - |