@inproceedings{zhao-etal-2025-explicit,
title = "Explicit vs. Implicit: Investigating Social Bias in Large Language Models through Self-Reflection",
author = "Zhao, Yachao and
Wang, Bo and
Wang, Yan and
Zhao, Dongming and
He, Ruifang and
Hou, Yuexian",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-acl.1/",
doi = "10.18653/v1/2025.findings-acl.1",
pages = "1--12",
ISBN = "979-8-89176-256-5",
abstract = "Large Language Models (LLMs) have been shown to exhibit various biases and stereotypes in their generated content. While extensive research has investigated biases in LLMs, prior work has predominantly focused on explicit bias, with minimal attention to implicit bias and the relation between these two forms of bias. This paper presents a systematic framework grounded in social psychology theories to investigate and compare explicit and implicit biases in LLMs.We propose a novel self-reflection-based evaluation framework that operates in two phases: first measuring implicit bias through simulated psychological assessment methods, then evaluating explicit bias by prompting LLMs to analyze their own generated content. Through extensive experiments on advanced LLMs across multiple social dimensions, we demonstrate that LLMs exhibit a substantial inconsistency between explicit and implicit biases: while explicit bias manifests as mild stereotypes, implicit bias exhibits strong stereotypes.We further investigate the underlying factors contributing to this explicit-implicit bias inconsistency, examining the effects of training data scale, model size, and alignment techniques. Experimental results indicate that while explicit bias declines with increased training data and model size, implicit bias exhibits a contrasting upward trend. Moreover, contemporary alignment methods effectively suppress explicit bias but show limited efficacy in mitigating implicit bias."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhao-etal-2025-explicit">
<titleInfo>
<title>Explicit vs. Implicit: Investigating Social Bias in Large Language Models through Self-Reflection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yachao</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bo</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yan</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dongming</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifang</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuexian</namePart>
<namePart type="family">Hou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wanxiang</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Nabende</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-256-5</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs) have been shown to exhibit various biases and stereotypes in their generated content. While extensive research has investigated biases in LLMs, prior work has predominantly focused on explicit bias, with minimal attention to implicit bias and the relation between these two forms of bias. This paper presents a systematic framework grounded in social psychology theories to investigate and compare explicit and implicit biases in LLMs.We propose a novel self-reflection-based evaluation framework that operates in two phases: first measuring implicit bias through simulated psychological assessment methods, then evaluating explicit bias by prompting LLMs to analyze their own generated content. Through extensive experiments on advanced LLMs across multiple social dimensions, we demonstrate that LLMs exhibit a substantial inconsistency between explicit and implicit biases: while explicit bias manifests as mild stereotypes, implicit bias exhibits strong stereotypes.We further investigate the underlying factors contributing to this explicit-implicit bias inconsistency, examining the effects of training data scale, model size, and alignment techniques. Experimental results indicate that while explicit bias declines with increased training data and model size, implicit bias exhibits a contrasting upward trend. Moreover, contemporary alignment methods effectively suppress explicit bias but show limited efficacy in mitigating implicit bias.</abstract>
<identifier type="citekey">zhao-etal-2025-explicit</identifier>
<identifier type="doi">10.18653/v1/2025.findings-acl.1</identifier>
<location>
<url>https://aclanthology.org/2025.findings-acl.1/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>1</start>
<end>12</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Explicit vs. Implicit: Investigating Social Bias in Large Language Models through Self-Reflection
%A Zhao, Yachao
%A Wang, Bo
%A Wang, Yan
%A Zhao, Dongming
%A He, Ruifang
%A Hou, Yuexian
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Findings of the Association for Computational Linguistics: ACL 2025
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-256-5
%F zhao-etal-2025-explicit
%X Large Language Models (LLMs) have been shown to exhibit various biases and stereotypes in their generated content. While extensive research has investigated biases in LLMs, prior work has predominantly focused on explicit bias, with minimal attention to implicit bias and the relation between these two forms of bias. This paper presents a systematic framework grounded in social psychology theories to investigate and compare explicit and implicit biases in LLMs.We propose a novel self-reflection-based evaluation framework that operates in two phases: first measuring implicit bias through simulated psychological assessment methods, then evaluating explicit bias by prompting LLMs to analyze their own generated content. Through extensive experiments on advanced LLMs across multiple social dimensions, we demonstrate that LLMs exhibit a substantial inconsistency between explicit and implicit biases: while explicit bias manifests as mild stereotypes, implicit bias exhibits strong stereotypes.We further investigate the underlying factors contributing to this explicit-implicit bias inconsistency, examining the effects of training data scale, model size, and alignment techniques. Experimental results indicate that while explicit bias declines with increased training data and model size, implicit bias exhibits a contrasting upward trend. Moreover, contemporary alignment methods effectively suppress explicit bias but show limited efficacy in mitigating implicit bias.
%R 10.18653/v1/2025.findings-acl.1
%U https://aclanthology.org/2025.findings-acl.1/
%U https://doi.org/10.18653/v1/2025.findings-acl.1
%P 1-12
Markdown (Informal)
[Explicit vs. Implicit: Investigating Social Bias in Large Language Models through Self-Reflection](https://aclanthology.org/2025.findings-acl.1/) (Zhao et al., Findings 2025)
ACL