FAIR-QR: Enhancing Fairness-aware Information Retrieval through Query Refinement
Fumian Chen, Hui Fang
TL;DR
This work addresses fairness in information retrieval, where search and recommendation systems can propagate underrepresented groups. It introduces FAIR-QR, a query-refinement framework that iteratively refines keywords to increase exposure of underrepresented groups, followed by relevance-preserving re-ranking. The approach is interpretable, since refined keywords are visible at each iteration, and demonstrates fairness gains (AWRF) while preserving strong relevance (nDCG) against baselines. It has practical impact for retrieval systems and retrieval-augmented generation pipelines, and is designed to generalize across retrievers and generative models with public code.
Abstract
Information retrieval systems such as open web search and recommendation systems are ubiquitous and significantly impact how people receive and consume online information. Previous research has shown the importance of fairness in information retrieval systems to combat the issue of echo chambers and mitigate the rich-get-richer effect. Therefore, various fairness-aware information retrieval methods have been proposed. Score-based fairness-aware information retrieval algorithms, focusing on statistical parity, are interpretable but could be mathematically infeasible and lack generalizability. In contrast, learning-to-rank-based fairness-aware information retrieval algorithms using fairness-aware loss functions demonstrate strong performance but lack interpretability. In this study, we proposed a novel and interpretable framework that recursively refines query keywords to retrieve documents from underrepresented groups and achieve group fairness. Retrieved documents using refined queries will be re-ranked to ensure relevance. Our method not only shows promising retrieval results regarding relevance and fairness but also preserves interpretability by showing refined keywords used at each iteration.