AI Paper Atlas

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

Activation Subspaces for Out-of-Distribution Detection

Barış Zöngür, Robin Hesse, Stefan Roth

ICCV 2025 · Label-efficient and robust learning › OOD and anomaly detection · approach: Provable and feature-based OOD detection · 5 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
add insignificant-subspace cosine score to decisive shaped-energy scoreSCALE-based decisive score · ImageNet-1k ResNet-50, 4 OOD setsAUC: 95.71 -> 96.74Improved
add ActSub to SCALESCALE · OpenOOD Near-OODAUC: 81.36 -> 84.24Improved
add ActSub to ASH-SASH-S · OpenOOD Near-OODAUC: 79.63 -> 84.06Improved
add ActSub to ReActReAct · OpenOOD Near-OODAUC: 77.38 -> 77.70Improved
add ActSub to SCALESCALE · CIFAR10 ID DenseNet-101AUC: 96.16 -> 96.87Improved
add ActSub to SCALESCALE · CIFAR100 ID DenseNet-101FPR@95TPR: 45.86 -> 41.26Improved
add ActSub to SCALESCALE · MobileNetV2 ImageNet-1kFPR@95TPR: 34.18 -> 21.70Improved
use SVD basis instead of PCA basis for the combined score (ablation)ActSub · OpenOOD Near-OODAUC: 80.40 -> 84.24Improved
use SVD basis instead of Null Space basis for the combined score (ablation)ActSub · OpenOOD Near-OODAUC: 80.04 -> 84.24Improved
use SVD basis instead of PCA for decisive-only score (ablation)decisive shaped score · OpenOOD Far-OODAUC: 96.10 -> 93.58Hurt
compute cosine score on insignificant component instead of full activation (ablation)insignificant cosine score · OpenOOD Far-OODAUC: 87.77 -> 92.74Improved
compute cosine score on insignificant component instead of full activation (ablation)insignificant cosine score · OpenOOD Near-OODAUC: 71.75 -> 67.51Hurt

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