Enhancing Homophily in Heterogeneous Graph Contrastive Learning via Connection Strength and Multi-view Self-Expression
Haosen Wang, Chenglong Shi, Can Xu, Surong Yan, Rong Xie, Pan Tang
Abstract
Heterogeneous graph pre-training (HGP) has demonstrated remarkable performance across various domains. However, the issue of heterophily in real-world heterogeneous graphs (HGs) has been largely neglected. To bridge this research gap, we proposed a novel heterogeneous graph contrastive learning framework, termed HGMS, which leverages connection strength and multi-view self-expression to learn homophilous node representations. Specifically, we design a heterogeneous edge dropping augmentation strategy, which preferentially preserves metapath-based edges with strong connection strength, thereby improving the homophily of augmented views. Moreover, we propose a multi-view self-expressive learning method to infer the homophily between nodes from the subspace. In practice, we develop two approaches to solve the self-expressive matrix. The solved self-expressive matrix serves as an additional augmented view to provide homophilous information and is used to mitigate false negatives in contrastive loss. Extensive experimental results demonstrate the superiority of HGMS across different downstream tasks. The code is available at https://github.com/senllh/HGMS/tree/main.
Ask Lune
Know where this paper stands in minutes, not an afternoon.
Lune can read the top-tier papers that cite it and the ones closest to it, and tells you what it claims, what came after and where it falls short, each point cited to the paper it comes from. No skimming, no AI slop.
1 top-tier paper has cited it since. Know what they found before your reviewers do.
Lune can draw on
its abstract · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
Cited by top-tier papers1
Related papers
- Graph Contrastive Learning via Interventional View GenerationZengyi Wo, Minglai Shao, Wenjun Wang, Xuan Guo et al.WWW 2024 · 13 citations
- Heterophily-aware Contrastive Learning for Heterophilic HypergraphsMing Li, Yongqi Li, Yuting Chen, Feilong Cao et al.AAAI 2026
- Self-Supervised Heterogeneous Graph Learning: a Homophily and Heterogeneity ViewYujie Mo, Feiping Nie, Ping Hu, Heng Tao Shen et al.ICLR 2024 · 17 citations
- HomoGCL: Rethinking Homophily in Graph Contrastive LearningWen-Zhi Li, Chang-Dong Wang, Hui Xiong, Jian-Huang LaiKDD 2023 · 31 citations
- Graph Self-supervised Learning with Augmentation-aware Contrastive LearningDong Chen, Xiang Zhao, Wei Wang, Zhen Tan et al.WWW 2023 · 17 citations
