AI Paper Atlas

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

PDD: Manifold-Prior Diverse Distillation for Medical Anomaly Detection

Xijun Lu, Hongying Liu, Fanhua Shang, Yanming Hui, Liang Wan

CVPR 2026 · Label-efficient and robust learning › OOD and anomaly detection · approach: Unified multi-class anomaly detection · 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
dual-teacher InA+MMU instead of single-teacher RD (ablation)PDD · ZhangLab Chest X-rayAUROC: 81.5 -> 90.8Improved
add MPA module (prior injection) to dual-teacher single student (ablation)PDD · ZhangLab Chest X-rayAUROC: 90.8 -> 92.9Improved
add MPA to single-teacher RD (ablation)RD (1t1s) · ZhangLab Chest X-rayAUROC: 81.5 -> 82.0Improved
upgrade to dual-student design (ablation)PDD · ZhangLab Chest X-rayAUROC: 92.9 -> 94.0Improved
use teacher-student cosine for anomaly map instead of student-student cosine (with Ldiv) (ablation)PDD · BrainMRIAUROC: 32.50 -> 96.67Improved
add diversity loss Ldiv (with teacher-student scoring) (ablation)PDD · BrainMRIAUROC: 93.41 -> 96.67Improved
vary tau_low/tau_high thresholds (ablation)PDD · HeadCTAUROC: 96.15 -> 97.31No clear effect
PDD vs strongest baseline SimSIDPDD · CheXpertAUROC: 79.7 -> 79.1No clear effect
PDD vs MambaAD on Uni-MedicalPDD · Uni-Medical (mean)F1 max: 82.0 -> 85.4Improved
PDD vs MambaAD on Uni-MedicalPDD · Uni-Medical (mean)AUROC: 83.7 -> 81.4Hurt
PDD vs Skip-TS on HeadCTPDD · HeadCTAUROC: 85.7 -> 97.5Improved

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