AI Paper Atlas

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

A Unified Interpretation of Training-Time Out-of-Distribution Detection

Xu Cheng, Xin Jiang, Zechao Li

ICCV 2025 · Label-efficient and robust learning › OOD and anomaly detection · approach: Provable and feature-based OOD detection · 0 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
penalize high-order interactions so model encodes mainly low-order ones (vs cross-entropy baseline) (ablation)ResNet-18 · CIFAR-10 ID, OOD averageFPR95: 62.03 -> 91.45Hurt
penalize high-order interactions so model encodes mainly low-order ones (ablation)ResNet-18 · CIFAR-10 ID, OOD averageAUROC: 88.48 -> 73.07Hurt
penalize high-order interactions so model encodes mainly low-order ones (ablation)ResNet-34 · CIFAR-100 ID, OOD averageAUROC: 78.30 -> 58.77Hurt
penalize high-order interactions so model encodes mainly low-order ones (ablation)WideResNet-40-2 · CIFAR-100 ID, OOD averageFPR95: 78.01 -> 90.33Hurt
penalize low-order interactions so model encodes mainly high-order ones (ablation)ResNet-18 · CIFAR-10 ID, OOD averageFPR95: 62.03 -> 53.05Improved
penalize low-order interactions so model encodes mainly high-order ones (ablation)ResNet-18 · CIFAR-100 ID, OOD averageAUROC: 78.15 -> 79.45Improved
penalize low-order interactions so model encodes mainly high-order ones (ablation)WideResNet-40-2 · CIFAR-10 ID, OOD averageFPR95: 56.99 -> 61.64Hurt
penalize low-order interactions so model encodes mainly high-order ones (ablation)ResNet-34 · CIFAR-100 ID, OOD averageFPR95: 78.95 -> 81.51Hurt
train with training-time OOD methods (CSI, LogitNorm, T2FNorm, DAL) instead of cross-entropy onlyResNet-18/34, WideResNet-40-2 · CIFAR-10 and CIFAR-100 IDrelative interaction strength: Delta R(m) > 0 for m > 0.75n, < 0 for m < 0.25nImproved
train with training-time OOD methods instead of cross-entropy onlyResNet-based models · near/far-OOD interaction similaritysimilarity of interaction distribution to IDImproved

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