RHGCL: Representation-Driven Hierarchical Graph Contrastive Learning for User-Item Recommendation

ICLR 2026 Conference Submission23205 Authors

Keywords: Item Representation, Recommendation, Contrastive Learning

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 Representation-driven Hierarchical Graph Contrastive Learning (RHGCL), a novel GCL method that incorporates hierarchical item structures from learned representations for recommendations. First, RHGCL pre-trains a GCL module using cross-layer contrastive learning to obtain user and item representations. Second, RHGCL employs a representation compression and clustering method to construct a two-hierarchy user-item bipartite graph. Ultimately, RHGCL 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 RHGCL over existing baseline models, highlighting the contribution of representation-driven hierarchical item structures in enhancing GCL methods for recommendation tasks.

Primary Area: unsupervised, self-supervised, semi-supervised, and supervised representation learning

Submission Number: 23205