AI Paper Atlas

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

GPFlow: Gaussian Prototype Probability Flow for Unsupervised Multi-Modal Anomaly Detection

Yiting Li, Xulei Yang, Jingyi Liao, Jing Zhang, Fayao 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 covariance-aware PMP router instead of attention-based prototype router (ablation)GPFlow · MVTec-3D-AD (10-shot)I-AUROC: 0.883 -> 0.935Improved
add cross-modal reconstruction to intra-modal (coupled design) (ablation)GPFlow · MVTec-3D-AD (10-shot)I-AUROC: 0.915 -> 0.935Improved
add instance-aware prior calibration (ablation)GPFlow · MVTec-3D-AD (10-shot)I-AUROC: 0.925 -> 0.935Improved
use Gaussian prototypes with PMP instead of VQ codebook with soft assignment (ablation)GPFlow · MVTec-3D-AD (10-shot)I-AUROC: 0.885 -> 0.935Improved
use VQ codebook with optimal transport instead of Gaussian + PMP (ablation)GPFlow · MVTec-3D-AD (10-shot)I-AUROC: 0.935 -> 0.751Hurt
use analytical PMP router instead of learned Schrodinger Bridge (DPDL) (ablation)GPFlow · MVTec-3D-AD (10-shot)I-AUROC: 0.908 -> 0.935Improved
use cross-modal reconstruction only instead of intra-modal only (ablation)GPFlow · MVTec-3D-AD (10-shot)I-AUROC: 0.915 -> 0.775Hurt
increase MSM selection ratio rho beyond 0.6 to 0.8 (ablation)GPFlow IPC · MVTec-3D-ADI-AUROC: 0.911 -> 0.907No clear effect
inject 100% anomalous patches into prototype calibration (ablation)GPFlow IPC · MVTec-3D-ADI-AUROC: 0.894 -> 0.872Hurt
inject up to 50% anomalous patches into prototype calibration (ablation)GPFlow IPC · MVTec-3D-ADI-AUROC: 0.894 -> 0.895Mixed
train with all normal data instead of few-shotGPFlow vs FIND · MVTec-3D-AD (all-shot)I-AUROC: 0.978 -> 0.970Hurt
use GPFlow instead of FIND in compute costGPFlow · EfficiencyFLOPs (G): 728.46 -> 129.27Improved

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