Out-of-Distribution Detection on ImageNet-1k vs iNaturalist

28개 결과 · ⬇ CSV · JSON

AUROC

Rank	Model	AUROC	FPR95	Latency, ms	Paper	Code	Year
1	Forte	99.67 ± 00.03	0.64 ± 00.06	–	Forte : Finding Outliers with Representation Typicality Estimation	DebarghaG/forte	2024
2	CMA(ViT-B/16, NegLabel)	99.62	1.65	–	Enhanced OoD Detection through Cross-Modal Alignment of Multi-Modal Representations	ma-kjh/CMA-OoDD	2025
3	NNGuide (RegNet)	99.57	1.83	31.00	Nearest Neighbor Guidance for Out-of-Distribution Detection	jingkang50/openood · roomo7time/nnguide	2023
4	ViM (ViT-B/16)	99.41	2.60	–	ViM: Out-Of-Distribution with Virtual-logit Matching	jingkang50/openood · haoqiwang/vim	2022
5	DisCoPatch	99.1	3.6	1.56	DisCoPatch: Taming Adversarially-driven Batch Statistics for Improved Out-of-Distribution Detection		2025
6	SCALE (ResNet50)	98.17	9.5	11.27	Scaling for Training Time and Post-hoc Out-of-distribution Detection Enhancement	kai422/scale	2023
7	MOS (BiT-S-R101x1)	98.15	9.28	–	MOS: Towards Scaling Out-of-distribution Detection for Large Semantic Space	deeplearning-wisc/large_scale_ood · deeplearning-wisc/gradnorm_ood · ma-kjh/CMA-OoDD · +2	2021
8	ASH-S (ResNet-50)	97.87	11.49	–	Extremely Simple Activation Shaping for Out-of-Distribution Detection	andrijazz/ash · litianliu/fdbd-ood	2022
9	NNGuide (ResNet50 w/ ReAct)	97.7	11.12	11.10	Nearest Neighbor Guidance for Out-of-Distribution Detection	jingkang50/openood · roomo7time/nnguide	2023
10	BATS (ResNet-50)	97.67	–	–	Boosting Out-of-distribution Detection with Typical Features		2022
11	LINe (ResNet-50)	97.56	12.26	–	LINe: Out-of-Distribution Detection by Leveraging Important Neurons	yonghyun-ahn/line-out-of-distribution-detection-by-leveraging-important-neurons	2023
12	NAC-UE (ResNet-50)	96.52	–	–	Neuron Activation Coverage: Rethinking Out-of-distribution Detection and Generalization	bierone/ood_coverage	2023
13	DICE + ReAct (ResNet-50)	96.24	18.64	–	DICE: Leveraging Sparsification for Out-of-Distribution Detection	deeplearning-wisc/dice	2021
14	NPOS	96.19	16.58	–	Non-Parametric Outlier Synthesis	deeplearning-wisc/npos	2023
15	MCM (CLIP-L)	94.95	28.38	–	Delving into Out-of-Distribution Detection with Vision-Language Representations	deeplearning-wisc/mcm · HHU-MMBS/plp-official-tmlr2024	2022
16	KNN (ResNet-50 SupCon)	94.72	30.83	–	Out-of-Distribution Detection with Deep Nearest Neighbors	deeplearning-wisc/knn-ood · ArthurHoa/deep-eknn · munhouiani/ood-knn	2022
17	ODIN+UMAP (ResNet-50)	94.71	21.97	–	Unleashing Mask: Explore the Intrinsic Out-of-Distribution Detection Capability	tmlr-group/unleashing-mask	2023
18	MCM (CLIP-B)	94.61	30.91	–	Delving into Out-of-Distribution Detection with Vision-Language Representations	deeplearning-wisc/mcm · HHU-MMBS/plp-official-tmlr2024	2022
19	RankFeat (ResNetv2-101)	91.91	41.31	–	RankFeat: Rank-1 Feature Removal for Out-of-distribution Detection	kingjamessong/rankfeat	2022
20	ReAct (ResNet-50)	91.53	42.40	–	ReAct: Out-of-distribution Detection With Rectified Activations	deeplearning-wisc/react	2021