HGCL: Hierarchical Graph Contrastive Learning for User-Item Recommendation

2025/05/25 by Jiawei Xue, Xue, Jiawei, Zhen Yang +11 · 1 citation
Computer Science · #Advanced Graph Neural Networks #Artificial Intelligence (cs.AI) #Expert finding and Q&A systems #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Recommender Systems and Techniques

paper · pdf · doi:10.48550/arxiv.2505.19020

openalex publication_date 2025/05/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Graph Contrastive Learning (GCL), which fuses graph neural networks with contrastive learning, has evolved as a pivotal tool in user-item recommendations. While promising, existing GCL methods often lack explicit modeling of hierarchical item structures, which represent item similarities across varying resolutions. Such hierarchical item structures are ubiquitous in various items (e.g., online products and local businesses), and reflect their inherent organizational properties that serve as critical signals for enhancing recommendation accuracy. In this paper, we propose Hierarchical Graph Contrastive Learning (HGCL), a novel GCL method that incorporates hierarchical item structures for user-item recommendations. First, HGCL pre-trains a GCL module using cross-layer contrastive learning to obtain user and item representations. Second, HGCL employs a representation compression and clustering method to construct a two-hierarchy user-item bipartite graph. Ultimately, HGCL fine-tunes user and item representations by learning on the hierarchical graph, and then provides recommendations based on user-item interaction scores. Experiments on three widely adopted benchmark datasets ranging from 70K to 382K nodes confirm the superior performance of HGCL over existing baseline models, highlighting the contribution of hierarchical item structures in enhancing GCL methods for recommendation tasks.

Citations

• OmniSage: Large Scale, Multi-Entity Heterogeneous Graph Representation Learning
• ContextGNN: Beyond Two-Tower Recommendation Systems
• Leveraging Contrastive Learning for Enhanced Node Representations in Tokenized Graph Transformers
• Towards Graph Contrastive Learning: A Survey and Beyond
• LiGNN: Graph Neural Networks at LinkedIn
• Macro Graph Neural Networks for Online Billion-Scale Recommender Systems
• Architecture Matters: Uncovering Implicit Mechanisms in Graph Contrastive Learning
• Simple and Asymmetric Graph Contrastive Learning without Augmentations
• Long-tail Augmented Graph Contrastive Learning for Recommendation
• LightGCL: Simple Yet Effective Graph Contrastive Learning for Recommendation
• XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation
• XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation
• Self-Supervised Learning for Recommender Systems: A Survey
• Representing Long-Range Context for Graph Neural Networks with Global Attention
• Are Graph Augmentations Necessary? Simple Graph Contrastive Learning for Recommendation
• InfoGCL: Information-Aware Graph Contrastive Learning
• Self-supervised Graph-level Representation Learning with Local and Global Structure
• Self-supervised Heterogeneous Graph Neural Network with Co-contrastive Learning
• Bootstrapping User and Item Representations for One-Class Collaborative Filtering
• Graph Contrastive Learning with Augmentations
• Deep Graph Contrastive Representation Learning
• A Simple Framework for Contrastive Learning of Visual Representations
• LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
• LightGCN
• Deep Graph Infomax
• Representation Learning with Contrastive Predictive Coding
• Hierarchical Graph Representation Learning with Differentiable Pooling
• BPR: Bayesian Personalized Ranking from Implicit Feedback

Cited by

• GR-LLMs: Recent Advances in Generative Recommendation Based on Large Language Models

Related

• LightGCL: Simple Yet Effective Graph Contrastive Learning for Recommendation
• Graph Contrastive Learning on Multi-label Classification for Recommendations
• NLGCL: Naturally Existing Neighbor Layers Graph Contrastive Learning for Recommendation
• XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation
• Hybrid Matrix Factorization Based Graph Contrastive Learning for Recommendation System