Hypergraph Contrastive Learning for both Homophilic and Heterophilic Hypergraphs

2025-11-24

AI Summary

Most existing hypergraph neural networks rely on the homophily assumption, limiting their ability to model real-world heterophilous hypergraphs. To address this, we propose HONOR, the first unsupervised contrastive learning framework tailored for heterophilous hypergraph representation learning. Methodologically, HONOR introduces three key innovations: (1) prompt-guided hyperedge feature construction, enabling globally semantically consistent high-order relational modeling; (2) an adaptive attention-based aggregation mechanism that dynamically captures node-specific contributions to hyperedges; and (3) high-pass filtering to enhance representation robustness and generalizability. Extensive experiments across diverse homophilous and heterophilous hypergraph benchmarks demonstrate that HONOR consistently outperforms state-of-the-art methods, validating its effectiveness, robustness to structural heterogeneity, and strong cross-scenario generalization capability.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Computer Vision: Representation Learning for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Hypergraphs, as a generalization of traditional graphs, naturally capture high-order relationships. In recent years, hypergraph neural networks (HNNs) have been widely used to capture complex high-order relationships. However, most existing hypergraph neural network methods inherently rely on the homophily assumption, which often does not hold in real-world scenarios that exhibit significant heterophilic structures. To address this limitation, we propose extbf{HONOR}, a novel unsupervised extbf{H}ypergraph c extbf{ON}trastive learning framework suitable for both hom extbf{O}philic and hete extbf{R}ophilic hypergraphs. Specifically, HONOR explicitly models the heterophilic relationships between hyperedges and nodes through two complementary mechanisms: a prompt-based hyperedge feature construction strategy that maintains global semantic consistency while suppressing local noise, and an adaptive attention aggregation module that dynamically captures the diverse local contributions of nodes to hyperedges. Combined with high-pass filtering, these designs enable HONOR to fully exploit heterophilic connection patterns, yielding more discriminative and robust node and hyperedge representations. Theoretically, we demonstrate the superior generalization ability and robustness of HONOR. Empirically, extensive experiments further validate that HONOR consistently outperforms state-of-the-art baselines under both homophilic and heterophilic datasets.

Problem

• Addressing hypergraph neural networks' reliance on homophily assumption
• Modeling heterophilic relationships between hyperedges and nodes
• Creating robust representations for both homophilic and heterophilic scenarios

Innovation

• Unsupervised hypergraph contrastive learning framework
• Prompt-based hyperedge feature construction strategy
• Adaptive attention aggregation with high-pass filtering