Papers › Detecting Out-of-distribution Data through In-distribution Class Prior
Detecting Out-of-distribution Data through In-distribution Class Prior
Xue Jiang, Feng Liu, Zhen Fang, Hong Chen, Tongliang Liu, Feng Zheng, Bo Han
Given a pre-trained in-distribution (ID) model, the inference-time out-of-distribution (OOD) detection aims to recognize OOD data during the inference stage. However, some representative methods share an unproven assumption that the probability that OOD data belong to every ID class should be the same, i.e., these OOD-to-ID probabilities actually form a uniform distribution. In this paper, we show that this assumption makes the above methods incapable when the ID model is trained with class-imbalanced data.Fortunately, by analyzing the causal relations between ID/OOD classes and features, we identify several common scenarios where the OOD-to-ID probabilities should be the ID-class-prior distribution and propose two strategies to modify existing inference-time detection methods: 1) replace the uniform distribution with the ID-class-prior distribution if they explicitly use the uniform distribution; 2) otherwise, reweight their scores according to the similarity between the ID-class-prior distribution and the softmax outputs of the pre-trained model. Extensive experiments show that both strategies can improve the OOD detection performance when the ID model is pre-trained with imbalanced data, reflecting the importance of ID-class prior in OOD detection.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Out-of-Distribution Detection | ImageNet-1k vs Curated OODs (avg.) | RP+GradNorm | FPR95 | 70.12 | #15 of 16 | Archive leaderboard | report |
| Out-of-Distribution Detection | ImageNet-1k vs Places | RP+GradNorm | FPR95 | 83.29 | #24 of 25 | Archive leaderboard | report |
| Out-of-Distribution Detection | ImageNet-1k vs SUN | RP+GradNorm | FPR95 | 73.53 | #20 of 22 | Archive leaderboard | report |
| Out-of-Distribution Detection | ImageNet-1k vs Textures | RP+GradNorm | FPR95 | 79.80 | #34 of 34 | Archive leaderboard | report |
| Out-of-Distribution Detection | ImageNet-1k vs iNaturalist | RP+GradNorm | FPR95 | 43.87 | #25 of 28 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections