AI Paper Atlas

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

Activation Matters: Test-time Activated Negative Labels for OOD Detection with Vision-Language Models

Yabin Zhang, Maya Varma, Yunhe Gao, Jean-Benoit Delbrouck, Jiaming Liu, Chong Wang, Curtis Langlotz

CVPR 2026 · Label-efficient and robust learning › OOD and anomaly detection · approach: Spurious-aware OOD detection · 2 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 fixed activation-agnostic negative labels with distribution-adaptive activated labels (ablation)NegLabel · ImageNet-1k OpenOOD Near-OODFPR95: 69.45 -> 61.57Improved
replace fixed activation-agnostic negative labels with distribution-adaptive activated labels (ablation)NegLabel · ImageNet-1k OpenOOD Far-OODFPR95: 23.73 -> 17.52Improved
add batch-adaptive activation and activation-aware score on top of distribution-adaptive labels (ablation)TANL · ImageNet-1k OpenOOD Near-OODFPR95: 61.57 -> 60.06Improved
add batch-adaptive activation and activation-aware score on top of distribution-adaptive labels (ablation)TANL · ImageNet-1k OpenOOD Far-OODFPR95: 17.52 -> 17.21No clear effect
use activation-aware score Saa instead of standard score Snl with many negative labels (ablation)TANL · ImageNet-1k OpenOODFPR95 vs label numberImproved
increase number of negative labels M (ablation)TANL · ImageNet-1k OpenOODFPR95Hurt
mine negative labels with negative-sample activation instead of positive information alone (ablation)TANL · ImageNet-1k OpenOODFPR95Improved
lower alpha from 1.0 to 0.95 to mix batch information (ablation)TANL · ImageNet-1k OpenOODFPR95Improved
lower alpha further below 0.95 (ablation)TANL · ImageNet-1k OpenOODFPR95Hurt
increase test batch size (ablation)TANL · ImageNet-1k OpenOODFPR95Improved
use TANL instead of NegLabelNegLabel · ImageNet-1k, 4 OOD datasetsFPR95: 25.40 -> 9.81Improved
use TANL instead of NegLabelNegLabel · Full-spectrum OpenOOD Near-OODFPR95: 76.25 -> 68.71Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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