AI Paper Atlas

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

Memory-Distilled Selection for Noise-Robust Anomaly Detection

Sirojbek Safarov, Jaewoo Park, Yoon Gyo Jung, Kuan-Chuan Peng, Wonchul Kim, Seongdeok Bang, Octavia Camps

ICML 2026 · Label-efficient and robust learning › OOD and anomaly detection · approach: Anomaly detection with contaminated data · 0 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 MeDS to DinomalyDinomaly · MVTecAD (40% noise)I-AUROC: 87.38 -> 99.16Improved
add MeDS to INP-FormerINP-Former · MVTecAD (40% noise)I-AUROC: 85.85 -> 99.17Improved
add MeDS to HVQHVQ · MVTecAD (40% noise)I-AUROC: 92.14 -> 94.26Improved
add MeDS to DinomalyDinomaly · MVTecAD (clean, 0% noise)I-AUROC: 99.64 -> 99.37Hurt
add MeDS to DinomalyDinomaly · MVTecAD (40% noise)P-AP: 53.00 -> 68.05Improved
add MeDS to DinomalyDinomaly · VisA (10% noise)I-AUROC: 93.56 -> 97.43Improved
add MeDS to DinomalyDinomaly · Real-IAD (40% noise, single-class)I-AUROC: 87.78 -> 90.99Improved
add bootstrapped memory ensemble to Dinomaly (ablation)Dinomaly · MVTecAD (40% noise), Dinomaly ablationI-AUROC: 90.45 -> 92.22Improved
add bootstrapped memory ensemble to Dinomaly (ablation)Dinomaly · MVTecAD (40% noise), Dinomaly ablationP-AP: 54.75 -> 53.04Hurt
add memory distillation on top of memory ensemble (ablation)Dinomaly · MVTecAD (40% noise), Dinomaly ablationI-AUROC: 92.22 -> 98.76Improved
add progressive-selection fine-tuning after distillation (random init, memory criteria) (ablation)Dinomaly · MVTecAD (40% noise), Dinomaly ablationP-AP: 56.88 -> 67.71Improved
initialize fine-tuning and select data with distilled model instead of random init (ablation)MeDS · MVTecAD (40% noise), Dinomaly ablationP-AP: 67.91 -> 68.05Improved

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