Learning to Shop Like Humans: A Review-driven Retrieval-Augmented Recommendation Framework with LLMs

2025-08-31

AI Summary

To address the low efficiency and poor relevance of dynamically integrating user reviews in LLM-based recommender systems—particularly the difficulty in focusing on decision-relevant reviews for the target item—this paper proposes RevBrowse. Its core innovation is the PrefRAG module, which decouples user and item representations and performs adaptive, fine-grained retrieval of preference-relevant reviews conditioned on the target item, followed by retrieval-augmented generation (RAG) for dynamic review re-ranking. This design jointly ensures contextual sensitivity, interpretability, and computational efficiency. Extensive experiments on four Amazon datasets demonstrate that RevBrowse significantly outperforms strong baselines, achieving superior performance in modeling dynamic user preferences and cross-domain generalization. Moreover, it provides traceable, review-grounded justifications for recommendations.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Learning Preferences or Rankings
• Knowledge Representation and Reasoning: Preferences

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• 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 shown strong potential in recommendation tasks due to their strengths in language understanding, reasoning and knowledge integration. These capabilities are especially beneficial for review-based recommendation, which relies on semantically rich user-generated texts to reveal fine-grained user preferences and item attributes. However, effectively incorporating reviews into LLM-based recommendation remains challenging due to (1) inefficient to dynamically utilize user reviews under LLMs' constrained context windows, and (2) lacking effective mechanisms to prioritize reviews most relevant to the user's current decision context. To address these challenges, we propose RevBrowse, a review-driven recommendation framework inspired by the "browse-then-decide" decision process commonly observed in online user behavior. RevBrowse integrates user reviews into the LLM-based reranking process to enhance its ability to distinguish between candidate items. To improve the relevance and efficiency of review usage, we introduce PrefRAG, a retrieval-augmented module that disentangles user and item representations into structured forms and adaptively retrieves preference-relevant content conditioned on the target item. Extensive experiments on four Amazon review datasets demonstrate that RevBrowse achieves consistent and significant improvements over strong baselines, highlighting its generalizability and effectiveness in modeling dynamic user preferences. Furthermore, since the retrieval-augmented process is transparent, RevBrowse offers a certain level of interpretability by making visible which reviews influence the final recommendation.

Problem

• Inefficient dynamic review usage under LLM context constraints
• Lacking mechanisms to prioritize decision-relevant user reviews
• Challenges in incorporating reviews into LLM-based recommendation systems

Innovation

• Retrieval-augmented module for adaptive review selection
• Browse-then-decide framework integrating reviews into reranking
• Structured user-item representation for transparent preference modeling