A Survey of Generalization of Graph Anomaly Detection: From Transfer Learning to Foundation Models

2025-09-08

AI Summary

Existing graph anomaly detection (GAD) methods typically assume identical training and testing distributions and are highly task-specific, leading to severely limited generalizability under realistic conditions such as distribution shift and few-shot settings. To address this, we propose the first systematic, fine-grained taxonomy for GAD generalization, unifying modeling across three dimensions: task formulation, types of distribution shift, and knowledge transfer mechanisms—spanning from traditional transfer learning to foundation models. We introduce the inaugural survey framework dedicated to GAD generalization, integrating state-of-the-art approaches for complex scenarios including cross-domain and low-resource settings. Our analysis explicitly identifies three core challenges: distribution shift correction, mitigation of label scarcity, and enhancement of model transferability. This work fills a critical gap in the literature by providing the first comprehensive, structured overview of GAD generalization, offering a clear technical roadmap and actionable directions for developing general-purpose graph anomaly detection systems.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Graph-based Machine Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Graph anomaly detection (GAD) has attracted increasing attention in recent years for identifying malicious samples in a wide range of graph-based applications, such as social media and e-commerce. However, most GAD methods assume identical training and testing distributions and are tailored to specific tasks, resulting in limited adaptability to real-world scenarios such as shifting data distributions and scarce training samples in new applications. To address the limitations, recent work has focused on improving the generalization capability of GAD models through transfer learning that leverages knowledge from related domains to enhance detection performance, or developing "one-for-all" GAD foundation models that generalize across multiple applications. Since a systematic understanding of generalization in GAD is still lacking, in this paper, we provide a comprehensive review of generalization in GAD. We first trace the evolution of generalization in GAD and formalize the problem settings, which further leads to our systematic taxonomy. Rooted in this fine-grained taxonomy, an up-to-date and comprehensive review is conducted for the existing generalized GAD methods. Finally, we identify current open challenges and suggest future directions to inspire future research in this emerging field.

Problem

• Addressing limited adaptability in graph anomaly detection
• Improving generalization across shifting data distributions
• Developing transfer learning and foundation models for GAD

Innovation

• Transfer learning for graph anomaly detection
• One-for-all foundation models approach
• Systematic taxonomy for generalization methods