AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

Rethinking Out-of-Distribution (OOD) Detection: Masked Image Modeling Is All You Need

Jingyao Li, Pengguang Chen, Zexin He, Shaozuo Yu, Shu Liu, Jiaya Jia

CVPR 2023 · Label-efficient and robust learning › OOD and anomaly detection · approach: Provable and feature-based OOD detection · 56 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Replace classification pretraining with MIM (ablation)ViT OOD detector · CIFAR-10 → CIFAR-100 OODAUROC: 98.6 -> 99.4Improved
Replace MoCov3 contrastive pretraining with MIM (ablation)ViT OOD detector · CIFAR-100 → CIFAR-10 OODAUROC: 72.8 -> 98.3Improved
Use MIM rather than MoCov3 pretraining (ablation)ViT OOD detector · CIFAR-10 one-classAUROC: 13.3% to 98.66%Improved
Add intermediate fine-tuning after MIM pretraining (ablation)MOOD · CIFAR-10 one-classAUROC: 72.2 -> 97.9Improved
Add intermediate fine-tuning after MIM pretraining (ablation)MOOD · CIFAR-100 one-classAUROC: 66.3 -> 96.5Improved
Add ID fine-tuning after MIM pretraining (ablation)MOOD · ImageNet-30 one-classAUROC: 75.2 -> 92.0Improved
Add both intermediate fine-tuning and ID fine-tuning (ablation)MOOD · CIFAR-10 multi-class OOD averageAUROC: 74.5 -> 99.7Improved
Use Mahalanobis distance instead of Softmax for OOD scoring (ablation)MOOD · CIFAR-10 → SVHN multi-class OODAUROC: 88.6 -> 99.8Improved
Use a single MIM-pretrained ViT instead of R50+ViT (ablation)OOD detector · CIFAR-100 ID, CIFAR-10 OODAUROC: 96.23 -> 98.30%Improved
Use MOOD without OOD exposure instead of R50+ViT trained with 10 OOD samples per classNear-distribution OOD detector · Near-distribution OOD detectionAUROC: 99.29% -> 99.41%Improved
Use MOOD rather than CSIOne-class OOD detection · CIFAR-10 one-class averageAUROC: +3.5Improved
Use MOOD rather than SSD+Multi-class OOD detection · CIFAR-10 OOD averageAUROC: +2.5Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy