RALLRec: Improving Retrieval Augmented Large Language Model Recommendation with Representation Learning

2025-02-10

AI Summary

Existing RAG-based recommender systems over-rely on textual semantics while neglecting user behavioral signals, leading to suboptimal retrieval relevance. To address this, we propose a collaborative-semantic joint-enhanced RAG framework. First, it jointly models textual and collaborative semantics via dual-channel integration: fine-grained item descriptions generated by large language models (LLMs) are fused with user–item interaction representations extracted from collaborative filtering. Second, a lightweight temporal-aware re-ranking module is introduced to explicitly capture dynamic user interest evolution. To the best of our knowledge, this is the first RAG recommendation method that enables end-to-end joint learning of LLM-derived textual semantics and collaborative representations. Extensive experiments on three real-world datasets demonstrate significant improvements—up to 12.7% higher Recall@10 over state-of-the-art methods. The source code is publicly available.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Large Language Models (LLMs) have been integrated into recommendation systems to enhance user behavior comprehension. The Retrieval Augmented Generation (RAG) technique is further incorporated into these systems to retrieve more relevant items and improve system performance. However, existing RAG methods rely primarily on textual semantics and often fail to incorporate the most relevant items, limiting the effectiveness of the systems. In this paper, we propose Representation learning for retrieval-Augmented Large Language model Recommendation (RALLRec). Specifically, we enhance textual semantics by prompting LLMs to generate more detailed item descriptions, followed by joint representation learning of textual and collaborative semantics, which are extracted by the LLM and recommendation models, respectively. Considering the potential time-varying characteristics of user interest, a simple yet effective reranking method is further introduced to capture the dynamics of user preference. We conducted extensive experiments on three real-world datasets, and the evaluation results validated the effectiveness of our method. Code is made public at https://github.com/JianXu95/RALLRec.

Problem

• Enhance recommendation systems with advanced representation learning.
• Improve retrieval of relevant items using collaborative and textual semantics.
• Capture dynamic user preferences with a novel reranking method.

Innovation

• Enhanced textual semantics via LLM prompting
• Joint representation learning of textual-collaborative semantics
• Reranking method for dynamic user preference capture