What question did this study set out to answer?

The central aim is to enhance search system fairness in information retrieval through improved retrieval techniques.

April 1, 2026Open Access

THUIR at the NTCIR-18 FairWeb-2 Task

Key Points

The central aim is to enhance search system fairness in information retrieval through improved retrieval techniques.
Developed a retrieval pipeline integrating neural rerankers
Aggregated results using reciprocal rank fusion
Combined PM2-based result diversification with dense retrieval scores
Conducted experiments to evaluate performance metrics.
Achieved competitive performance across multiple evaluation metrics
Showed a positive relationship between retrieval relevance and group fairness
Proved that a combination of techniques enhances both relevance and fairness effectively.

Abstract

The fairness of search systems remains a critical challenge in information retrieval. Building upon our previous work in FairWeb‑1, this paper presents the THUIR team’s approach in the NTCIR‑18 FairWeb‑2 Task. Specifically, we developed a simple yet effective retrieval pipeline that integrates multiple neural rerankers with results aggregated via Reciprocal Rank Fusion to generate balanced search rankings across various entity types. Additionally, we submitted a revived run that combines a PM2-based result diversification algorithm with dense retrieval scores. Our experimental results yield competitive performance on multiple evaluation metrics, demonstrating that enhancements in retrieval relevance inherently promote balanced group fairness. With the right combination of techniques, it is possible to achieve a synergistic reinforcement between relevance and fairness.

Demander à l'IA

Bookmark

View Full Paper