DCGL: Dual-Channel Graph Learning with Large Language Models for Knowledge-Aware Recommendation

2026-05-08

AI Summary

This work addresses key limitations in existing knowledge graph–enhanced recommendation methods that integrate large language models, which often fail to effectively model implicit semantic relationships, suffer from interference between embedding channels, and overlook variations in user interaction frequency. To overcome these issues, the authors propose a dual-channel graph learning framework that decouples and separately models semantic information and user behavioral signals. The framework incorporates multi-level contrastive learning to enhance representation robustness and introduces a dynamic fusion mechanism weighted by interaction frequency to adaptively integrate information from both channels. Extensive experiments on four real-world datasets demonstrate that the proposed method significantly outperforms state-of-the-art approaches, with particularly notable gains under data sparsity while maintaining high recommendation accuracy for active users.

Abstract

Knowledge Graphs (KGs) have proven highly effective for recommendation systems by capturing latent item relationships, while recent integration of Large Language Models (LLMs) has further enhanced semantic understanding and addressed knowledge sparsity issues. Nevertheless, current KG-and-LLM-based methods still face three main limitations: 1) inadequate modeling of implicit semantic relationships beyond explicit KG links; 2) suboptimal single-channel fusion of ID and LLM embeddings, which often leads to signal interference and blurred representations; and 3) insufficient consideration of user-item interaction frequency variations in recommendation strategies. To address these challenges, we propose the Dual-Channel Graph Learning (DCGL) framework, featuring three key innovations: 1) a dual-channel architecture that structurally decouples rich semantic information from user behavioral patterns, preventing early interference; 2) a multi-level contrastive learning mechanism that enhances robustness against KG noise through intra-view contrasts and bridges semantic gaps between channels via inter-view alignment; and 3) a dynamic fusion mechanism that adaptively balances semantic generalization and behavioral specificity based on interaction frequency, resolving the cascading limitation. Extensive experiments on four real-world datasets show that DCGL consistently outperforms state-of-the-art methods, yielding substantial improvements in sparse scenarios while maintaining precision for active users. Our code is available at https://github.com/XinchiZou/DCGL.

Problem

• Knowledge Graph
• Large Language Models
• Recommendation Systems
• Semantic Relationships
• Interaction Frequency

Innovation

• Dual-Channel Architecture
• Contrastive Learning
• Dynamic Fusion
• Knowledge Graph
• Large Language Models