Generalized Pseudo-Relevance Feedback DOI: XXXXXXX.XXXXXXX Conference: Make sure to enter the correct conference title from your rights confirmation email; June 03–05, 2018; Woodstock, NY ISBN: 978-1-4503-XXXX-X/18/06 CCS: Information systems Query reformulation CCS: Computing methodologies Natural

Yiteng Tu, Weihang Su, Yujia Zhou, Yiqun Liu, Fen Lin, Qin Liu, Qingyao Ai · 2025

Read Generalized Pseudo-Relevance Feedback DOI: XXXXXXX.XXXXXXX Conference: Make sure to enter the correct conference title from your rights confirmation email; June 03–05, 2018; Woodstock, NY ISBN: 978-1-4503-XXXX-X/18/06 CCS: Information systems Query reformulation CCS: Computing methodologies Natural by Yiteng Tu, Weihang Su, Yujia Zhou, Yiqun Liu, Fen Lin, Qin Liu, Qingyao Ai (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-25488.txt.

Description

Yiteng Tu led six co-authors on pseudo-relevance feedback. Weihang Su, Yujia Zhou, and Yiqun Liu helped. Fen Lin, Qin Liu, and Qingyao Ai joined. Their 2025 paper generalizes query rewriting for search. Query rewriting refines user queries using retrieval results. Classic feedback assumes top documents are relevant.

Classic methods also assume certain model architectures for rewriting. Generative feedback with language models relaxes architecture needs. Their Generalized Pseudo-Relevance Feedback relaxes both assumptions. It performs model-free natural language rewriting from documents. Dependence on the relevance assumption is reduced greatly. A utility-oriented training pipeline uses reinforcement learning. That ensures robustness against noisy feedback signals. Extensive tests covered multiple benchmarks and retrievers. Their approach always outperformed strong rivals there. Search systems may benefit across varied retriever types.

Catalog record

CREATORYiteng Tu, Weihang Su, Yujia Zhou, Yiqun Liu, Fen Lin, Qin Liu, Qingyao Ai
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE50 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2510-25488

Browse the full library · About the Tower of Babel