Enhancing Homophily-Heterophily Separation: Relation-Aware Learning in Heterogeneous Graphs

2025-06-25

AI Summary

Modeling node disassortativity in real-world heterogeneous graphs faces dual challenges: jointly capturing both node/edge heterogeneity and disassortative connectivity patterns. Existing approaches often coarsely homogenize heterogeneous graphs, leading to loss of disassortative signals. To address this, we propose RASH, a relation-aware assortative–disassortative separation framework. RASH constructs a doubly heterogeneous hypergraph to encode multi-relational subgraphs and explicitly decouples assortative and disassortative patterns—first achieved in heterogeneous graph learning. It dynamically generates dual views based on relation importance and introduces a multi-relational contrastive loss to align heterogeneous semantics with assortative/disassortative structures. Furthermore, high-order semantic fusion is enhanced via multi-relational mutual information maximization. Extensive experiments on multiple benchmark datasets demonstrate significant improvements in node classification and link prediction, validating RASH’s effectiveness and generalizability in modeling complex disassortative structures.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Real-world networks usually have a property of node heterophily, that is, the connected nodes usually have different features or different labels. This heterophily issue has been extensively studied in homogeneous graphs but remains under-explored in heterogeneous graphs, where there are multiple types of nodes and edges. Capturing node heterophily in heterogeneous graphs is very challenging since both node/edge heterogeneity and node heterophily should be carefully taken into consideration. Existing methods typically convert heterogeneous graphs into homogeneous ones to learn node heterophily, which will inevitably lose the potential heterophily conveyed by heterogeneous relations. To bridge this gap, we propose Relation-Aware Separation of Homophily and Heterophily (RASH), a novel contrastive learning framework that explicitly models high-order semantics of heterogeneous interactions and adaptively separates homophilic and heterophilic patterns. Particularly, RASH introduces dual heterogeneous hypergraphs to encode multi-relational bipartite subgraphs and dynamically constructs homophilic graphs and heterophilic graphs based on relation importance. A multi-relation contrastive loss is designed to align heterogeneous and homophilic/heterophilic views by maximizing mutual information. In this way, RASH simultaneously resolves the challenges of heterogeneity and heterophily in heterogeneous graphs. Extensive experiments on benchmark datasets demonstrate the effectiveness of RASH across various downstream tasks. The code is available at: https://github.com/zhengziyu77/RASH.

Problem

• Addresses node heterophily in heterogeneous graphs
• Models high-order semantics of heterogeneous interactions
• Separates homophilic and heterophilic patterns adaptively

Innovation

• Dual heterogeneous hypergraphs encode multi-relational subgraphs
• Dynamic homophilic and heterophilic graphs construction
• Multi-relation contrastive loss aligns heterogeneous views