Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Evolutionary Router Feature Generation for Zero-Shot Graph Anomaly Detection with Mixture-of-Experts

Haiyang Jiang, Tong Chen, Xinyi Gao, Guansong Pang, Quoc Viet Hung Nguyen, Hongzhi Yin

TL;DR

A novel MoE framework with evolutionary router feature generation (EvoFG) for zero-shot GAD and an evolutionary feature generation scheme that iteratively constructs and selects informative structural features via an LLM-based generator and Shapley-guided evaluation are proposed.

Abstract

Zero-shot graph anomaly detection (GAD) has attracted increasing attention recent years, yet the heterogeneity of graph structures, features, and anomaly patterns across graphs make existing single GNN methods insufficiently expressive to model diverse anomaly mechanisms. In this regard, Mixture-of-experts (MoE) architectures provide a promising paradigm by integrating diverse GNN experts with complementary inductive biases, yet their effectiveness in zero-shot GAD is severely constrained by distribution shifts, leading to two key routing challenges. First, nodes often carry vastly different semantics across graphs, and straightforwardly performing routing based on their features is prone to generating biased or suboptimal expert assignments. Second, as anomalous graphs often exhibit pronounced distributional discrepancies, existing router designs fall short in capturing domain-invariant routing principles that generalize beyond the training graphs. To address these challenges, we propose a novel MoE framework with evolutionary router feature generation (EvoFG) for zero-shot GAD. To enhance MoE routing, we propose an evolutionary feature generation scheme that iteratively constructs and selects informative structural features via an LLM-based generator and Shapley-guided evaluation. Moreover, a memory-enhanced router with an invariant learning objective is designed to capture transferable routing patterns under distribution shifts. Extensive experiments on six benchmarks show that EvoFG consistently outperforms state-of-the-art baselines, achieving strong and stable zero-shot GAD performance.

Evolutionary Router Feature Generation for Zero-Shot Graph Anomaly Detection with Mixture-of-Experts

TL;DR

A novel MoE framework with evolutionary router feature generation (EvoFG) for zero-shot GAD and an evolutionary feature generation scheme that iteratively constructs and selects informative structural features via an LLM-based generator and Shapley-guided evaluation are proposed.

Abstract

Zero-shot graph anomaly detection (GAD) has attracted increasing attention recent years, yet the heterogeneity of graph structures, features, and anomaly patterns across graphs make existing single GNN methods insufficiently expressive to model diverse anomaly mechanisms. In this regard, Mixture-of-experts (MoE) architectures provide a promising paradigm by integrating diverse GNN experts with complementary inductive biases, yet their effectiveness in zero-shot GAD is severely constrained by distribution shifts, leading to two key routing challenges. First, nodes often carry vastly different semantics across graphs, and straightforwardly performing routing based on their features is prone to generating biased or suboptimal expert assignments. Second, as anomalous graphs often exhibit pronounced distributional discrepancies, existing router designs fall short in capturing domain-invariant routing principles that generalize beyond the training graphs. To address these challenges, we propose a novel MoE framework with evolutionary router feature generation (EvoFG) for zero-shot GAD. To enhance MoE routing, we propose an evolutionary feature generation scheme that iteratively constructs and selects informative structural features via an LLM-based generator and Shapley-guided evaluation. Moreover, a memory-enhanced router with an invariant learning objective is designed to capture transferable routing patterns under distribution shifts. Extensive experiments on six benchmarks show that EvoFG consistently outperforms state-of-the-art baselines, achieving strong and stable zero-shot GAD performance.
Paper Structure (37 sections, 23 equations, 6 figures, 6 tables, 2 algorithms)

This paper contains 37 sections, 23 equations, 6 figures, 6 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Graph Anomaly Detection
  4. Problem Formulation
  5. Methodology
  6. GNN Mixture-of-Experts
  7. LLM-Based Router Feature Generation
  8. Router Feature Selection via Efficient Shapley Estimation
  9. Memory-enhanced Router
  10. Invariant Routing Learning
  11. Overall Training Pipeline
  12. Experiments
  13. Experimental Setup
  14. Zero-shot Performance (Q1)
  15. Ablation Study (Q2)
  16. ...and 22 more sections

Figures (6)

  • Figure 1: Overall framework of EvoFG. Top: the iterative training pipeline. Bottom: detailed workflow.
  • Figure 2: Hyper-parameter analysis of feature generation.
  • Figure 3: Soft routing heatmap across datasets.
  • Figure 4: Coefficient $\lambda$ of invariant learning in EvoFG (Full) across different datasets.
  • Figure 5: Number of environments of invariant learning in EvoFG (Full) across different datasets.
  • ...and 1 more figures