Research map / Label-efficient and robust learning / OOD and anomaly detection
Is Task-Specific Training Necessary for Anomaly Detection?
Xingwu Zhang, Guanxuan Li, Paul Henderson, Gerardo Aragon-Camarasa, ZIJUN LONG
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add global [CLS] retrieval before patch matching (ablation) | RAD · MVTec-AD | inference latency: 293.7 ms -> 44.7 ms | Improved |
| compress memory bank with coreset sampling (ablation) | RAD · MVTec-AD | pixel AP: 75.60 -> 73.86 | Hurt |
| compress memory bank with coreset sampling (ablation) | RAD · MVTec-AD | pixel AUPRO: 94.90 -> 94.75 | No clear effect |
| increase anomaly-free data ratio from 0.05 to 1.00 (single-class) | RAD · MVTec-AD | image AUROC: 97.1 -> 99.6 | Improved |
| increase anomaly-free data ratio from 0.05 to 1.00 (single-class) | RAD · MVTec-AD | pixel AP: 70.6 -> 75.7 | Improved |
| use retrieval instead of trained decoder at tau=0.05 (single-class) | RAD vs Dinomaly · MVTec-AD | pixel AP: 54.7 -> 70.6 | Improved |
| use retrieval instead of trained decoder at tau=0.05 (multi-class) | RAD vs Dinomaly · MVTec-AD | pixel AP: 51.1 -> 67.9 | Improved |
| use retrieval instead of trained decoder with 2 transistor images | RAD vs Dinomaly · MVTec-AD incremental | pixel AP: 33.1 -> 53.7 | Improved |
| use retrieval instead of trained decoder | RAD vs Dinomaly · MVTec-AD few-shot | pixel AUPRO: +16.8 / +16.9 / +14.9 at Y=1,2,4 | Improved |
| use RAD instead of few-shot method IIPAD | RAD vs IIPAD · MVTec-AD few-shot | pixel AUPRO: +2.1 / +2.7 / +2.7 at Y=1,2,4 | Improved |
| reduce full data to a single anomaly-free image (ablation) | RAD · MVTec-AD | pixel AUROC: 98.5 -> 96.7 | Hurt |
| swap in stronger frozen encoder (DINOv3 vs MIM encoders) (ablation) | RAD · MVTec-AD | P-AUPRO | Improved |
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.
- Dinomaly — Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection (CVPR 2025) · better on 14 of 16 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
- ResAD: A Simple Framework for Class Generalizable Anomaly Detection (NeurIPS 2024 · similarity 0.89)
- Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection (CVPR 2025 · similarity 0.87)
- Generating and Reweighting Dense Contrastive Patterns for Unsupervised Anomaly Detection (AAAI 2024 · similarity 0.87)
- Long-Tailed Anomaly Detection with Learnable Class Names (CVPR 2024 · similarity 0.87)
- One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection (NeurIPS 2024 · similarity 0.86)
- Normal-Abnormal Guided Generalist Anomaly Detection (NeurIPS 2025 · similarity 0.86)
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.
