Graph Anomaly Detection Comparison Hub

3 papers - avg viability 5.0

Reference Surfaces

TA-GGAD: Testing-time Adaptive Graph Model for Generalist Graph Anomaly Detection(8.0)
A novel graph foundation model that achieves state-of-the-art cross-domain anomaly detection in graph data.
Mitigating Homophily Disparity in Graph Anomaly Detection: A Scalable and Adaptive Approach(7.0)
SAGAD is a scalable graph anomaly detection framework that adaptively fuses low- and high-frequency information to identify anomalous nodes in large-scale graphs, offering superior accuracy and scalability.
AC2L-GAD: Active Counterfactual Contrastive Learning for Graph Anomaly Detection(5.0)
An innovative framework using active counterfactual contrastive learning to enhance graph anomaly detection efficiency and performance.
Balanced Anomaly-guided Ego-graph Diffusion Model for Inductive Graph Anomaly Detection(5.0)
A dynamic graph anomaly detection model designed to improve generalization in fraud and cybersecurity applications by synthesizing balanced anomaly patterns.
Interpretable Graph-Level Anomaly Detection via Contrast with Normal Prototypes(5.0)
ProtoGLAD offers interpretable graph-level anomaly detection by contrasting graph anomalies with nearest normal prototypes.