New method improves recommendation queries using retrieval feedback
R2QRec: Learning Recommendation Queries from Retrieval Feedback
Information Retrieval
Summary
Recommendation systems suggest items you might like, but it's tricky to turn past user choices into search queries that find those items again. The authors present R2QRec, a method that learns to create good search queries without needing labeled examples of queries. It uses a fixed search system to check which queries find the right items, then improves queries based on how well they rank those items. This approach helps the system get better recommendations across different datasets without changing the search engine itself.
What this means in practice
- •For e-commerce developers: Enhance product recommendations by generating better search queries from user interaction data without needing query annotations.$Commercial implications: Improves commercial recommendation engines by refining query generation, leading to better item discovery and increased sales.
- •For digital library managers: Improve search and recommendation results for digital collections by learning query formulations from user behavior without altering the retrieval system.
Authors
Weixin Chen, Shuchang Liu, Hailan Yang, Yuhan Zhao, Li Kang, Xuefeng Zhang, Chenghao Zhang, Xinyuan Wang, Hanchen Yang, Yinghao Wu, Hua Huang, Yandong Bai, Fan Yang, Xiang Li, Chunjie Chen, Li Chen
Abstract
Query-based recommendation separates expressing user needs from locating matching items. Its central learning problem is how to turn item interactions into effective retrieval requests when interaction logs do not annotate queries. We present R2QRec, which uses a fixed retrieval system to construct query supervision and evaluate recommendation outcomes. Retrieval verification first selects generated expressions that recover observed items as supervised fine tuning targets, allowing retrieval behavior to determine the target wording. Policy refinement then evaluates generated query sets by the observed item's rank in their fused recommendation list. Log-rank feedback distinguishes outcomes beyond the recommendation cutoff, while a KL penalty constrains policy change relative to the supervised reference. The stages have distinct roles: verification identifies expressions to imitate, and policy refinement improves the combined ranking produced by newly generated queries. Only the query generator is trained; the retriever and item index remain fixed, and no query annotations or learned reward model are required. Across three public datasets, R2QRec achieves state-of-the-art overall recommendation performance among the evaluated methods. Extensive experiments demonstrate the effectiveness of the proposed design and its generalization across generators and domains.