Modeling Spectral Energy Shifts in Spatio-Temporal Graph Anomaly Detection

2026-05-29

AI Summary

This work addresses the challenge of detecting stealthy anomalies in graphs that masquerade as normal nodes by exhibiting abnormally reduced spectral energy. To this end, the authors propose a node-level spectral energy modeling approach that unifies the characterization of spectral energy shifts in both static and dynamic graphs through an energy-driven message-passing mechanism. Notably, this is the first method to explicitly incorporate spectral energy depletion into the anomaly detection framework. The proposed energy-aware architecture is inherently compatible with message passing and naturally extends to long-horizon dynamic graph settings without requiring specialized temporal modules. Experimental results demonstrate that the method significantly improves detection performance and scalability for concealed anomalies across multiple large-scale benchmark datasets.

Abstract

Graph anomaly detection methods aim to distinguish anomalous nodes. While prior methods characterize anomalies through increased variation in the spectral energy distributions, they overlook those that result in decreased variation, i.e., camouflaged anomalies that appear normal. We show that this type of anomaly persists across multiple datasets and remains undetectable by existing spectral approaches. To address this limitation, we propose a node-level spectral energy formulation that is fully compatible with message passing and enables the detection of camouflaged anomalies. Building on this formulation, we introduce an energy-aware graph learning framework that models spectral shifts through energy-driven message passing in both static and time-series graphs. Besides, our unified architecture extends to temporal settings without introducing specialized sequence modules, enabling efficient learning under long sliding windows. Extensive experiments on large-scale benchmarks demonstrate the effectiveness and scalability of our approach.

Problem

• graph anomaly detection
• spectral energy
• camouflaged anomalies
• spatio-temporal graphs
• anomaly detection

Innovation

• spectral energy shift
• camouflaged anomaly
• energy-aware graph learning
• message passing
• spatio-temporal graph