<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ohi-etal-2024-likelihood">
    <titleInfo>
        <title>Likelihood-based Mitigation of Evaluation Bias in Large Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Masanari</namePart>
        <namePart type="family">Ohi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Masahiro</namePart>
        <namePart type="family">Kaneko</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ryuto</namePart>
        <namePart type="family">Koike</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Mengsay</namePart>
        <namePart type="family">Loem</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Naoaki</namePart>
        <namePart type="family">Okazaki</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2024-08</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics ACL 2024</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Lun-Wei</namePart>
            <namePart type="family">Ku</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Andre</namePart>
            <namePart type="family">Martins</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Vivek</namePart>
            <namePart type="family">Srikumar</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Bangkok, Thailand and virtual meeting</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Large Language Models (LLMs) are widely used to evaluate natural language generation tasks as automated metrics.However, the likelihood, a measure of LLM’s plausibility for a sentence, can vary due to superficial differences in sentences, such as word order and sentence structure.It is therefore possible that there might be a likelihood bias if LLMs are used for evaluation: they might overrate sentences with higher likelihoods while underrating those with lower likelihoods.In this paper, we investigate the presence and impact of likelihood bias in LLM-based evaluators.We also propose a method to mitigate the likelihood bias.Our method utilizes highly biased instances as few-shot examples for in-context learning.Our experiments in evaluating the data-to-text and grammatical error correction tasks reveal that several LLMs we test display a likelihood bias.Furthermore, our proposed method successfully mitigates this bias, also improving evaluation performance (in terms of correlation of models with human scores) significantly.</abstract>
    <identifier type="citekey">ohi-etal-2024-likelihood</identifier>
    <location>
        <url>https://aclanthology.org/2024.findings-acl.193</url>
    </location>
    <part>
        <date>2024-08</date>
        <extent unit="page">
            <start>3237</start>
            <end>3245</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Likelihood-based Mitigation of Evaluation Bias in Large Language Models
%A Ohi, Masanari
%A Kaneko, Masahiro
%A Koike, Ryuto
%A Loem, Mengsay
%A Okazaki, Naoaki
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand and virtual meeting
%F ohi-etal-2024-likelihood
%X Large Language Models (LLMs) are widely used to evaluate natural language generation tasks as automated metrics.However, the likelihood, a measure of LLM’s plausibility for a sentence, can vary due to superficial differences in sentences, such as word order and sentence structure.It is therefore possible that there might be a likelihood bias if LLMs are used for evaluation: they might overrate sentences with higher likelihoods while underrating those with lower likelihoods.In this paper, we investigate the presence and impact of likelihood bias in LLM-based evaluators.We also propose a method to mitigate the likelihood bias.Our method utilizes highly biased instances as few-shot examples for in-context learning.Our experiments in evaluating the data-to-text and grammatical error correction tasks reveal that several LLMs we test display a likelihood bias.Furthermore, our proposed method successfully mitigates this bias, also improving evaluation performance (in terms of correlation of models with human scores) significantly.
%U https://aclanthology.org/2024.findings-acl.193
%P 3237-3245

Markdown (Informal)

[Likelihood-based Mitigation of Evaluation Bias in Large Language Models](https://aclanthology.org/2024.findings-acl.193) (Ohi et al., Findings 2024)

• Likelihood-based Mitigation of Evaluation Bias in Large Language Models (Ohi et al., Findings 2024)

ACL

• Masanari Ohi, Masahiro Kaneko, Ryuto Koike, Mengsay Loem, and Naoaki Okazaki. 2024. Likelihood-based Mitigation of Evaluation Bias in Large Language Models. In Findings of the Association for Computational Linguistics ACL 2024, pages 3237–3245, Bangkok, Thailand and virtual meeting. Association for Computational Linguistics.