Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

RaDAR: Relation-aware Diffusion-Asymmetric Graph Contrastive Learning for Recommendation

Yixuan Huang, Jiawei Chen, Shengfan Zhang, Zongsheng Cao

Abstract

Collaborative filtering (CF) recommendation has been significantly advanced by integrating Graph Neural Networks (GNNs) and Graph Contrastive Learning (GCL). However, (i) random edge perturbations often distort critical structural signals and degrade semantic consistency across augmented views, and (ii) data sparsity hampers the propagation of collaborative signals, limiting generalization. To tackle these challenges, we propose RaDAR (Relation-aware Diffusion-Asymmetric Graph Contrastive Learning Framework for Recommendation Systems), a novel framework that combines two complementary view generation mechanisms: a graph generative model to capture global structure and a relation-aware denoising model to refine noisy edges. RaDAR introduces three key innovations: (1) asymmetric contrastive learning with global negative sampling to maintain semantic alignment while suppressing noise; (2) diffusion-guided augmentation, which employs progressive noise injection and denoising for enhanced robustness; and (3) relation-aware edge refinement, dynamically adjusting edge weights based on latent node semantics. Extensive experiments on three public benchmarks demonstrate that RaDAR consistently outperforms state-of-the-art methods, particularly under noisy and sparse conditions.

RaDAR: Relation-aware Diffusion-Asymmetric Graph Contrastive Learning for Recommendation

Abstract

Collaborative filtering (CF) recommendation has been significantly advanced by integrating Graph Neural Networks (GNNs) and Graph Contrastive Learning (GCL). However, (i) random edge perturbations often distort critical structural signals and degrade semantic consistency across augmented views, and (ii) data sparsity hampers the propagation of collaborative signals, limiting generalization. To tackle these challenges, we propose RaDAR (Relation-aware Diffusion-Asymmetric Graph Contrastive Learning Framework for Recommendation Systems), a novel framework that combines two complementary view generation mechanisms: a graph generative model to capture global structure and a relation-aware denoising model to refine noisy edges. RaDAR introduces three key innovations: (1) asymmetric contrastive learning with global negative sampling to maintain semantic alignment while suppressing noise; (2) diffusion-guided augmentation, which employs progressive noise injection and denoising for enhanced robustness; and (3) relation-aware edge refinement, dynamically adjusting edge weights based on latent node semantics. Extensive experiments on three public benchmarks demonstrate that RaDAR consistently outperforms state-of-the-art methods, particularly under noisy and sparse conditions.
Paper Structure (44 sections, 27 equations, 5 figures, 6 tables)

This paper contains 44 sections, 27 equations, 5 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Preliminaries and Related Work
  3. Collaborative Filtering Paradigm
  4. Self-Supervised Graph Learning
  5. Methodology
  6. User-item Embedding Propagation
  7. GCL Paradigm
  8. Graph Generative Model as View Generator
  9. Relation-Aware Graph Denoising for View Generation
  10. Diffusion with User-Item Graph
  11. Noise Diffusion Process
  12. Diffusion Process Optimization for User-Item Interaction.
  13. Contrastive Learning paradigms
  14. Diffusion-Enhanced Graph Contrastive Learning
  15. Asymmetric Graph Contrastive Learning
  16. ...and 29 more sections

Figures (5)

  • Figure 1: ACL and diffusion model on a user–item graph, illustrating how standard diffusion misses two-hop monophily where indirectly connected users share similar preferences.
  • Figure 2: RaDAR framework architecture: The left section shows two view generators extracting complementary graph representations. The right section demonstrates the contrastive learning process with diffusion model-based graph generation and joint optimization through InfoNCE, IB, and BPR losses.
  • Figure 3: Impact of Noise Ratio (5%--25%) on Performance Degradation
  • Figure 4: Performance variation with ACL ratio $\lambda$. Last.FM peaks Recall@20 at $\lambda=5.5$, NDCG@20 at $\lambda=3.5$. Yelp peaks Recall@20 at $\lambda=1.5$, NDCG@20 at $\lambda=1.0$. Higher $\lambda$ values enhance relation-aware denoising for Last.FM, while Yelp requires balanced contributions due to interaction sparsity.
  • Figure 5: Performance analysis across five user and item interaction sparsity levels on Yelp dataset.