AI Paper Atlas

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

FastRef: Fast Prototype Refinement for Few-shot Industrial Anomaly Detection

Yufei Li, Long Tian, Yuyang Dai, Wenchao Chen, Liang Bao, Xiyang Liu

CVPR 2026 · Label-efficient and robust learning › OOD and anomaly detection · approach: Unified multi-class anomaly detection

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 FastRef prototype refinementPatchCore · VisA 2-shotimage AUROC: 80.0 -> 87.1Improved
add FastRefPatchCore · MPDD 2-shotimage AUROC: 71.4 -> 78.2Improved
add FastRefWinCLIP · MVTec 2-shotimage AUROC: 93.7 -> 93.9No clear effect
add FastRefWinCLIP · MVTec 2-shotpixel AUROC: 93.8 -> 96.2Improved
add FastRefAnomalyDINO · MVTec 2-shotimage AUROC: 96.7 -> 97.2No clear effect
add FastRefAnomalyDINO · MPDD 2-shotimage AUROC: 75.2 -> 78.4Improved
add FastRefWinCLIP · MPDD 4-shotimage AUROC: 75.0 -> 82.0Improved
add FastRefPatchCore · MPDD 4-shot pixelpixel AUROC: 97.2 -> 97.2No clear effect
add characteristic transfer W* (ablation)WinCLIP+ · MPDD 2-shotimage AUROC: 72.5 -> 74.9Improved
add anomaly suppression T* on top of W* (ablation)WinCLIP+ · MPDD 2-shotimage AUROC: 74.9 -> 76.0Improved
add anomaly suppression T* (ablation)WinCLIP+ · MVTec/VisA/MPDD 2-shotimage AUROC: 93.8 -> 93.9; 84.0 -> 84.1; 74.9 -> 76.0Improved
use OT regularizer instead of mean-prototype Gaussian assumption (FastRecon form) (ablation)FastRef · MPDD 2-shotimage AUROC: 76.4 -> 78.2Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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