RALLREC+: Retrieval augmented large language model recommendation with reasoning

Large Language Models (LLMs) have been integrated into recommender 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 have two shortcomings. (i) In the retrieval stage, they rely primarily on textual semantics and often fail to incorporate the most relevant items, thus constraining system effectiveness. (ii) In the generation stage, they lack explicit chain-of-thought reasoning, further limiting their potential.
In this paper, we propose Representation learning and Reasoning empowered retrieval-Augmented Large Language model Recommendation (RALLREC+). Specifically, for the retrieval stage, we prompt LLMs to generate detailed item descriptions and perform joint representation learning, combining textual and collaborative signals extracted from the LLM and recommendation models, respectively. To account for the time-varying nature of user interests, we propose a simple yet effective reranking method to capture preference dynamics. For the generation phase, we first evaluate reasoning LLMs on recommendation tasks, uncovering valuable insights. Then we introduce knowledge-injected prompting and consistency-based merging approach to integrate reasoning LLMs with general-purpose LLMs, enhancing overall performance. Extensive experiments on three real-world datasets validate our method's effectiveness.
© 2025 Elsevier Ltd