Rethinking Feature Alignment in Generalist Graph Anomaly Detection: A Relational Fingerprint-based Approach

2026-05-25

AI Summary

This work addresses the challenge of negative transfer in existing general-purpose graph anomaly detection methods when applied across different graphs, primarily caused by the neglect of feature semantics and the resulting inability to learn transferable semantic knowledge. To overcome this limitation, the authors propose a novel Relation Fingerprint (ReFi)-based framework for universal graph anomaly detection. The approach introduces semantic-aware relation fingerprints—replacing conventional PCA alignment—to effectively preserve anomaly-relevant contextual and structural cues. It further integrates a Transformer encoder with a signal-to-noise ratio (SNR)-guided domain adaptation module to jointly capture domain-invariant and domain-specific knowledge. Notably, the framework enables cross-graph anomaly detection without requiring retraining on target graphs. Extensive experiments across 14 datasets demonstrate that the proposed method significantly outperforms state-of-the-art models, confirming its superior generalization capability and detection performance.

Abstract

Generalist graph anomaly detection (GAD) aims to detect anomalies on unseen graphs without graph-specific retraining. Nevertheless, existing approaches primarily focus on aligning heterogeneous features across different data domains via PCA-based projection, which harmonizes feature dimensions ignores feature semantics. As a result, GAD models fail to learn transferable semantic knowledge, and even exhibit negative transfer on unseen graphs. To address this issue, we propose a Relational Fingerprint-based generalist GAD approach (ReFi-GAD for short), aligning heterogeneous raw features with a universal and semantics-aware Relational Fingerprint (ReFi) that encodes anomaly-indicative cues from both contextual and structural perspectives. Building on ReFi, we design a fingerprint-grounded generalist GAD model, which combines a transformer-based encoder to capture domain-invariant knowledge with an SNR-guided refinement module for domain-specific adaptation. Extensive experiments on 14 datasets demonstrate that ReFi-GAD significantly outperforms state-of-the-art methods.

Problem

• Generalist Graph Anomaly Detection
• Feature Alignment
• Feature Semantics
• Negative Transfer
• Heterogeneous Features

Innovation

• Relational Fingerprint
• Generalist Graph Anomaly Detection
• Feature Alignment
• Semantic-aware Representation
• Domain Adaptation