<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="fan-etal-2026-fairsynth">
    <titleInfo>
        <title>FairSynth: Instrument-Relative Evaluation and Calibration of LLM Survey Simulations for Measurement Fairness</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Guangrui</namePart>
        <namePart type="family">Fan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">DanDan</namePart>
        <namePart type="family">Liu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Qian</namePart>
        <namePart type="family">Guo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xueying</namePart>
        <namePart type="family">Niu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lihu</namePart>
        <namePart type="family">Pan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-462-0</identifier>
    </relatedItem>
    <abstract>Large language models (LLMs) are increasingly used as virtual survey respondents, yet most evaluations compare only item-level marginal response distributions. Such evaluations can miss two failures central to computational social science: synthetic respondents may distort correlations among attitudes, and errors may be distributed unevenly across countries or demographic groups. We introduce FairSynth, an instrument-relative framework for auditing and calibrating closed-ended, answer-only LLM survey simulations against existing anchor surveys. FairSynth evaluates marginal fidelity, Structural Attitude Fidelity (item–item correlation similarity), group representation error, distributional flattening, and between-group sign reversals. It then applies a model-agnostic post-hoc logit-shift layer with shared group parameters to reduce disparities in representation error while preserving attitude structure. Experiments on World Values Survey Wave 7 and General Social Survey 2022 show that FairSynth reduces worst-group representation error and sign reversals while improving structural fidelity on the evaluated questionnaires. FairSynth is not a substitute for human surveys; it is a measurement-fairness toolkit for auditing, stress-testing, and calibrating synthetic survey simulations when anchor data are available.</abstract>
    <identifier type="citekey">fan-etal-2026-fairsynth</identifier>
    <identifier type="doi">10.18653/2026.findings-emnlp.714</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.714/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>14992</start>
            <end>15012</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T FairSynth: Instrument-Relative Evaluation and Calibration of LLM Survey Simulations for Measurement Fairness
%A Fan, Guangrui
%A Liu, DanDan
%A Guo, Qian
%A Niu, Xueying
%A Pan, Lihu
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F fan-etal-2026-fairsynth
%X Large language models (LLMs) are increasingly used as virtual survey respondents, yet most evaluations compare only item-level marginal response distributions. Such evaluations can miss two failures central to computational social science: synthetic respondents may distort correlations among attitudes, and errors may be distributed unevenly across countries or demographic groups. We introduce FairSynth, an instrument-relative framework for auditing and calibrating closed-ended, answer-only LLM survey simulations against existing anchor surveys. FairSynth evaluates marginal fidelity, Structural Attitude Fidelity (item–item correlation similarity), group representation error, distributional flattening, and between-group sign reversals. It then applies a model-agnostic post-hoc logit-shift layer with shared group parameters to reduce disparities in representation error while preserving attitude structure. Experiments on World Values Survey Wave 7 and General Social Survey 2022 show that FairSynth reduces worst-group representation error and sign reversals while improving structural fidelity on the evaluated questionnaires. FairSynth is not a substitute for human surveys; it is a measurement-fairness toolkit for auditing, stress-testing, and calibrating synthetic survey simulations when anchor data are available.
%R 10.18653/2026.findings-emnlp.714
%U https://aclanthology.org/2026.findings-emnlp.714/
%U https://doi.org/10.18653/2026.findings-emnlp.714
%P 14992-15012

Markdown (Informal)

[FairSynth: Instrument-Relative Evaluation and Calibration of LLM Survey Simulations for Measurement Fairness](https://aclanthology.org/2026.findings-emnlp.714/) (Fan et al., Findings 2026)

• FairSynth: Instrument-Relative Evaluation and Calibration of LLM Survey Simulations for Measurement Fairness (Fan et al., Findings 2026)

ACL

• Guangrui Fan, DanDan Liu, Qian Guo, Xueying Niu, and Lihu Pan. 2026. FairSynth: Instrument-Relative Evaluation and Calibration of LLM Survey Simulations for Measurement Fairness. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 14992–15012, Budapest, Hungary. Association for Computational Linguistics.