Out Of Distribution Detection On Imagenet 1K 8

AUROC

FPR95

Résultats

Résultats de performance de divers modèles sur ce benchmark

Tableau comparatif

Nom du modèle	AUROC	FPR95
rankfeat-rank-1-feature-removal-for-out-of	94.07	29.27
non-parametric-outlier-synthesis	90.44	43.77
out-of-distribution-detection-based-on-in	-	54.19
on-the-effectiveness-of-sparsification-for	93.94	25.45
react-out-of-distribution-detection-with-1	88.16	47.69
out-of-distribution-detection-with-deep	88.40	48.91
mos-towards-scaling-out-of-distribution	92.01	40.63
delving-into-out-of-distribution-detection	94.14	29.00
nearest-neighbor-guidance-for-out-of-1	94.43	21.58
unleashing-mask-explore-the-intrinsic-out-of	86.92	49.69
delving-into-out-of-distribution-detection	92.57	37.59
on-the-effectiveness-of-sparsification-for	90.83	35.15
extremely-simple-activation-shaping-for-out	94.02	27.98
rethinking-out-of-distribution-ood-detection	89.8	-
scaling-for-training-time-and-post-hoc-out-of	95.02	23.27
line-out-of-distribution-detection-by	95.26	19.48
on-the-importance-of-gradients-for-detecting	-	46.48
out-of-distribution-detection-with-deep	80.10	69.53
out-of-distribution-detection-with-implicit	76.26	80.94
decoupling-maxlogit-for-out-of-distribution	-	57.40
detecting-out-of-distribution-data-through-in	-	73.53