List your company
ResearchFormerly WizWand

Graph Anomaly Detection on YelpChi (AUC, F1-Macro)

95.84AUCSignGAD
53.002464.123775.24586.3663May 26, 2026
Updated 5mo ago

Evaluation results

MethodLinks
SignGADTraining Ratio=10%2026.0595.8486.52
SignGADTraining Ratio=5%2026.0595.2785.86
SignGADTraining Ratio=3%2026.0594.1384.31
SignGADTraining Ratio=1%2026.0592.6881.55
SignGADtraining ratio=1%, LLM backbone=GPT-4.12026.0592.6881.55
TAQ-GADTraining Ratio=10%2026.0591.8381.16
TAQ-GADTraining Ratio=5%2026.0591.0680.24
TAQ-GADTraining Ratio=3%2026.0589.5778.63
ConsisGADTraining Ratio=10%2026.0589.2676.32
SpaceGNNTraining Ratio=10%2026.0588.6477.57
ConsisGADTraining Ratio=5%2026.0588.1375.14
SpaceGNNTraining Ratio=5%2026.0587.5676.21
TAQ-GADTraining Ratio=1%2026.0586.9575.49
TAQ-GADtraining ratio=1%2026.0586.9575.49
ConsisGADTraining Ratio=3%2026.0586.5173.27
SpaceGNNTraining Ratio=3%2026.0585.8374.08
ConsisGADTraining Ratio=1%2026.0583.3669.72
ConsisGADtraining ratio=1%2026.0583.3669.72
SpaceGNNTraining Ratio=1%2026.0582.4770.21
SpaceGNNtraining ratio=1%2026.0582.4770.21
GraphConsistraining ratio=1%2026.0580.1364.96
BWGNNtraining ratio=1%2026.0577.6266.54
GDNtraining ratio=1%2026.0575.9264.81
GHRNtraining ratio=1%2026.0575.3363.62
PC-GNNtraining ratio=1%2026.0575.1764.23
DiG-In-GNNtraining ratio=1%2026.0574.463.36
GraphSAGEtraining ratio=1%2026.0573.763.33
GATv2training ratio=1%2026.0572.8362.23
H2-Fdetectortraining ratio=1%2026.0572.3863.97
MLPtraining ratio=1%2026.0572.1861.61
CARE-GNNtraining ratio=1%2026.0572.1161.62
GAGAtraining ratio=1%2026.0571.6161.81
GATtraining ratio=1%2026.0570.1461.22
GINtraining ratio=1%2026.0556.9853.58
GCNtraining ratio=1%2026.0554.6535.59