Can LLM Annotations Replace User Clicks for Learning to Rank?

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yu, Lulu, Bi, Keping, Guo, Jiafeng, Liu, Shihao, Wang, Shuaiqiang, Yin, Dawei, Cheng, Xueqi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918193302339584
author Yu, Lulu
Bi, Keping
Guo, Jiafeng
Liu, Shihao
Wang, Shuaiqiang
Yin, Dawei
Cheng, Xueqi
author_facet Yu, Lulu
Bi, Keping
Guo, Jiafeng
Liu, Shihao
Wang, Shuaiqiang
Yin, Dawei
Cheng, Xueqi
contents Large-scale supervised data is essential for training modern ranking models, but obtaining high-quality human annotations is costly. Click data has been widely used as a low-cost alternative, and with recent advances in large language models (LLMs), LLM-based relevance annotation has emerged as another promising annotation. This paper investigates whether LLM annotations can replace click data for learning to rank (LTR) by conducting a comprehensive comparison across multiple dimensions. Experiments on both a public dataset, TianGong-ST, and an industrial dataset, Baidu-Click, show that click-supervised models perform better on high-frequency queries, while LLM annotation-supervised models are more effective on medium- and low-frequency queries. Further analysis shows that click-supervised models are better at capturing document-level signals such as authority or quality, while LLM annotation-supervised models are more effective at modeling semantic matching between queries and documents and at distinguishing relevant from non-relevant documents. Motivated by these observations, we explore two training strategies -- data scheduling and frequency-aware multi-objective learning -- that integrate both supervision signals. Both approaches enhance ranking performance across queries at all frequency levels, with the latter being more effective. Our code is available at https://github.com/Trustworthy-Information-Access/LLMAnn_Click.
format Preprint
id arxiv_https___arxiv_org_abs_2511_06635
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Can LLM Annotations Replace User Clicks for Learning to Rank?
Yu, Lulu
Bi, Keping
Guo, Jiafeng
Liu, Shihao
Wang, Shuaiqiang
Yin, Dawei
Cheng, Xueqi
Information Retrieval
Large-scale supervised data is essential for training modern ranking models, but obtaining high-quality human annotations is costly. Click data has been widely used as a low-cost alternative, and with recent advances in large language models (LLMs), LLM-based relevance annotation has emerged as another promising annotation. This paper investigates whether LLM annotations can replace click data for learning to rank (LTR) by conducting a comprehensive comparison across multiple dimensions. Experiments on both a public dataset, TianGong-ST, and an industrial dataset, Baidu-Click, show that click-supervised models perform better on high-frequency queries, while LLM annotation-supervised models are more effective on medium- and low-frequency queries. Further analysis shows that click-supervised models are better at capturing document-level signals such as authority or quality, while LLM annotation-supervised models are more effective at modeling semantic matching between queries and documents and at distinguishing relevant from non-relevant documents. Motivated by these observations, we explore two training strategies -- data scheduling and frequency-aware multi-objective learning -- that integrate both supervision signals. Both approaches enhance ranking performance across queries at all frequency levels, with the latter being more effective. Our code is available at https://github.com/Trustworthy-Information-Access/LLMAnn_Click.
title Can LLM Annotations Replace User Clicks for Learning to Rank?
topic Information Retrieval
url https://arxiv.org/abs/2511.06635