HeTa: Relation-wise Heterogeneous Graph Foundation Attack Model

2025-06-09

AI Summary

This work addresses the challenge of robustness evaluation for heterogeneous graph neural networks (HGNNs) by proposing the first relation-aware universal attack foundation model. Unlike prior methods, it generates transferable structural perturbations across diverse HGNN architectures and heterogeneous graphs without retraining. Methodologically, we identify a common vulnerability pattern of HGNNs along the relational dimension—previously unobserved—and design a relation-level shared attack unit mining mechanism. We further introduce a sequential relation-weighted attack paradigm integrating a lightweight surrogate model, relation importance alignment, and heterogeneity normalization. Experiments demonstrate high attack success rates across multiple HGNN architectures and heterogeneous graph benchmarks. The generated perturbations exhibit strong cross-model transferability, and adaptation to new graphs requires only a few fine-tuning steps. This significantly enhances both black-box attack efficiency and generalization capability.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Heterogeneous Graph Neural Networks (HGNNs) are vulnerable, highlighting the need for tailored attacks to assess their robustness and ensure security. However, existing HGNN attacks often require complex retraining of parameters to generate specific perturbations for new scenarios. Recently, foundation models have opened new horizons for the generalization of graph neural networks by capturing shared semantics across various graph distributions. This leads us to ask:Can we design a foundation attack model for HGNNs that enables generalizable perturbations across different HGNNs, and quickly adapts to new heterogeneous graphs (HGs)? Empirical findings reveal that, despite significant differences in model design and parameter space, different HGNNs surprisingly share common vulnerability patterns from a relation-aware perspective. Therefore, we explore how to design foundation HGNN attack criteria by mining shared attack units. In this paper, we propose a novel relation-wise heterogeneous graph foundation attack model, HeTa. We introduce a foundation surrogate model to align heterogeneity and identify the importance of shared relation-aware attack units. Building on this, we implement a serialized relation-by-relation attack based on the identified relational weights. In this way, the perturbation can be transferred to various target HGNNs and easily fine-tuned for new HGs. Extensive experiments exhibit powerful attack performances and generalizability of our method.

Problem

• Assessing HGNN robustness via generalizable attack model
• Overcoming complex retraining for scenario-specific perturbations
• Exploiting shared vulnerability patterns across diverse HGNNs

Innovation

• Foundation surrogate model aligns heterogeneity
• Serialized relation-by-relation attack implementation
• Mining shared relation-aware attack units