Transferable Hypergraph Attack via Injecting Nodes into Pivotal Hyperedges

2025-11-12

AI Summary

Existing hypergraph neural network (HGNN) adversarial attack methods rely on target model architecture and overlook a common vulnerability arising from heterogeneous hyperedge importance, thereby limiting transferability and generalization. This paper proposes the first transferable adversarial attack framework targeting hyperedge importance: (1) a hyperedge identifier locates critical hyperedges; (2) a feature inversion mechanism generates malicious nodes inducing maximal semantic shift; and (3) a model-agnostic injection strategy executes the attack without access to target model parameters. The core innovation lies in the first application of hyperedge importance assessment to cross-model transfer attacks, with perturbation generation explicitly driven by semantic divergence maximization. Evaluated on six real-world datasets, our method significantly outperforms state-of-the-art approaches—particularly under cross-model transfer settings—demonstrating superior robustness degradation and attack transferability.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Multiagent Systems: Adversarial Agents

Application Category

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

Abstract

Recent studies have demonstrated that hypergraph neural networks (HGNNs) are susceptible to adversarial attacks. However, existing methods rely on the specific information mechanisms of target HGNNs, overlooking the common vulnerability caused by the significant differences in hyperedge pivotality along aggregation paths in most HGNNs, thereby limiting the transferability and effectiveness of attacks. In this paper, we present a novel framework, i.e., Transferable Hypergraph Attack via Injecting Nodes into Pivotal Hyperedges (TH-Attack), to address these limitations. Specifically, we design a hyperedge recognizer via pivotality assessment to obtain pivotal hyperedges within the aggregation paths of HGNNs. Furthermore, we introduce a feature inverter based on pivotal hyperedges, which generates malicious nodes by maximizing the semantic divergence between the generated features and the pivotal hyperedges features. Lastly, by injecting these malicious nodes into the pivotal hyperedges, TH-Attack improves the transferability and effectiveness of attacks. Extensive experiments are conducted on six authentic datasets to validate the effectiveness of TH-Attack and the corresponding superiority to state-of-the-art methods.

Problem

• Attacking hypergraph neural networks via adversarial node injection
• Improving attack transferability across different hypergraph models
• Targeting pivotal hyperedges to enhance attack effectiveness

Innovation

• Injecting malicious nodes into pivotal hyperedges
• Assessing hyperedge pivotality via recognizer module
• Generating features by maximizing semantic divergence