Mitigating the Threshold Priming Effect in Large Language Model-Based Relevance Judgments via Personality Infusing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Nuo, Fang, Hanpei, Liu, Jiqun, Wei, Wilson, Sakai, Tetsuya, Wu, Xiao-Ming
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908681497477120
author Chen, Nuo
Fang, Hanpei
Liu, Jiqun
Wei, Wilson
Sakai, Tetsuya
Wu, Xiao-Ming
author_facet Chen, Nuo
Fang, Hanpei
Liu, Jiqun
Wei, Wilson
Sakai, Tetsuya
Wu, Xiao-Ming
contents Recent research has explored LLMs as scalable tools for relevance labeling, but studies indicate they are susceptible to priming effects, where prior relevance judgments influence later ones. Although psychological theories link personality traits to such biases, it is unclear whether simulated personalities in LLMs exhibit similar effects. We investigate how Big Five personality profiles in LLMs influence priming in relevance labeling, using multiple LLMs on TREC 2021 and 2022 Deep Learning Track datasets. Our results show that certain profiles, such as High Openness and Low Neuroticism, consistently reduce priming susceptibility. Additionally, the most effective personality in mitigating priming may vary across models and task types. Based on these findings, we propose personality prompting as a method to mitigate threshold priming, connecting psychological evidence with LLM-based evaluation practices.
format Preprint
id arxiv_https___arxiv_org_abs_2512_00390
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Mitigating the Threshold Priming Effect in Large Language Model-Based Relevance Judgments via Personality Infusing
Chen, Nuo
Fang, Hanpei
Liu, Jiqun
Wei, Wilson
Sakai, Tetsuya
Wu, Xiao-Ming
Computation and Language
Information Retrieval
Recent research has explored LLMs as scalable tools for relevance labeling, but studies indicate they are susceptible to priming effects, where prior relevance judgments influence later ones. Although psychological theories link personality traits to such biases, it is unclear whether simulated personalities in LLMs exhibit similar effects. We investigate how Big Five personality profiles in LLMs influence priming in relevance labeling, using multiple LLMs on TREC 2021 and 2022 Deep Learning Track datasets. Our results show that certain profiles, such as High Openness and Low Neuroticism, consistently reduce priming susceptibility. Additionally, the most effective personality in mitigating priming may vary across models and task types. Based on these findings, we propose personality prompting as a method to mitigate threshold priming, connecting psychological evidence with LLM-based evaluation practices.
title Mitigating the Threshold Priming Effect in Large Language Model-Based Relevance Judgments via Personality Infusing
topic Computation and Language
Information Retrieval
url https://arxiv.org/abs/2512.00390