AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Fine-grained Abnormality Prompt Learning for Zero-shot Anomaly Detection

Jiawen Zhu, Yew-Soon Ong, Chunhua Shen, Guansong Pang

ICCV 2025 · Vision-language and multimodal › CLIP and zero-shot · approach: Attribute-based zero-shot learning · 28 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 CAP (compound abnormality prompts) to AnomalyCLIP (ablation)AnomalyCLIP · Industrial avg image-levelAUROC: 85.0 -> 88.1Improved
add orthogonal constraint loss to CAP (ablation)CAP · Industrial avg image-levelAUROC: 87.2 -> 88.1Improved
add DAP (data-dependent abnormality prior) to AnomalyCLIP (ablation)AnomalyCLIP · Medical avg image-levelAUROC: 87.7 -> 90.2Improved
add abnormality prior selection loss to DAP (ablation)DAP · Industrial avg image-levelAUROC: 86.5 -> 86.9Improved
combine CAP and DAP (ablation)AnomalyCLIP · Industrial avg image-levelAUROC: 88.1 -> 88.5Improved
use abnormal patch token prior (P) instead of image token prior (I) (ablation)FAPrompt · Industrial avg image-levelAUROC: 87.3 -> 88.5Improved
replace CoCoOp image-token prior with FAPrompt components (Lprior, CAP) (ablation)CoCoOp · Industrial avg image-levelAUROC: 73.1 -> 87.3Improved
use CAP prompts instead of ensembling AnomalyCLIP with different seeds (ablation)AnomalyCLIP Ensemble · Industrial avg image-levelAUROC: 85.5 -> 88.5Improved
average abnormality prompts instead of best-matched prompt selection (ablation)FAPrompt · Industrial avg image-levelAUROC: 85.1 -> 88.5Improved
average prompts instead of learned weighted aggregation (ablation)FAPrompt · Industrial avg image-levelAUROC: 87.7 -> 88.5Improved
increase number of abnormality prompts K beyond 10 (ablation)FAPrompt · Industrial datasetsAUROCMixed
increase number of selected patch tokens M beyond 10 (ablation)FAPrompt · Industrial datasetsAUROCMixed

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