AI Paper Atlas

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

MRAD: Zero-Shot Anomaly Detection with Memory-Driven Retrieval

Chaoran Xu, Chengkan Lv, Qiyu Chen, Feng Zhang, Zhengtao Zhang

ICLR 2026 · Label-efficient and robust learning › OOD and anomaly detection · approach: Unified multi-class anomaly detection · 6 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 lightweight two-linear-layer retrieval fine-tuning (ablation)MRAD-TF · 16-dataset averageI-AUROC: 81.0 -> 92.0Improved
add lightweight two-linear-layer retrieval fine-tuning (ablation)MRAD-TF · 16-dataset averageP-AUROC: 85.5 -> 91.9Improved
add prior-guided dynamic prompt biases (MRAD-CLIP) (ablation)MRAD-FT · 16-dataset averagePRO: 78.3 -> 80.1Improved
add prior-guided dynamic prompt biases (MRAD-CLIP) (ablation)MRAD-FT · 16-dataset averageI-AUROC: 92.0 -> 92.7Improved
bias prompts with CLIP class token instead of static prompts (ablation)MRAD-CLIP · MVTec-ADP-AUROC: 91.5 -> 86.2Hurt
add cross-patch attention bias to static prompts (ablation)MRAD-CLIP · MVTec-ADP-AUROC: 91.5 -> 92.5Improved
add dual normal+anomalous prior bias instead of anomaly prior only (ablation)MRAD-CLIP · Average over datasets pixel-levelP-AUROC: 92.0 -> 92.7Improved
add dual prior bias instead of cross-patch bias (ablation)MRAD-CLIP · Average over datasets image-levelI-AUROC: 92.1 -> 92.7Improved
use BTAD instead of VisA as auxiliary dataset (ablation)MRAD-CLIP · MVTec-ADP-AUROC: 93.0 -> 91.0Hurt
use SDD instead of VisA as auxiliary dataset (ablation)MRAD-CLIP · MVTec-ADI-AUROC: 94.0 -> 87.8Hurt
add image-level memory to pixel-level-only memory (ablation)MRAD · image-level detectionI-AUROCImproved
add pixel-level memory to image-level-only memory (ablation)MRAD · image-level detectionI-AUROCImproved

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