<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zheng-etal-2026-vispa">
    <titleInfo>
        <title>VISPA: Pluralistic Alignment via Automatic Value Selection and Activation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Shenyan</namePart>
        <namePart type="family">Zheng</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jiayou</namePart>
        <namePart type="family">Zhong</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Anudeex</namePart>
        <namePart type="family">Shetty</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Heng</namePart>
        <namePart type="family">Ji</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Preslav</namePart>
        <namePart type="family">Nakov</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Usman</namePart>
        <namePart type="family">Naseem</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>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</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">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>As large language models are increasingly deployed in high-stakes domains, it is essential that their outputs reflect not average human preference, but diverse perspectives. Achieving such pluralism remains challenging: existing pluralistic alignment methods rely on prompt-level interventions, represent only a limited set of values, or are compute-heavy, lacking explicit value control and representation. Furthermore, existing steering methods require manual value selection and cannot dynamically adapt to scenario-dependent value requirements, limiting their scalability and generality. To address these limitations, we introduce VISPA, a training-free pluralistic alignment framework that performs automatic, context-driven value selection and modular value-steered generation. VISPA identifies scenario-relevant values dynamically and applies activation steering to produce value-conditioned responses representing distinct perspectives. These responses enable support for three pluralistic alignment modes: Overton, Steerable, and Distributional. VISPA supports the addition of new values without finetuning and is compatible with diverse steering instantiations and steering models. Across extensive experiments spanning multiple models and evaluation settings, VISPA demonstrates strong performance across all pluralistic alignment modes in healthcare and beyond.</abstract>
    <identifier type="citekey">zheng-etal-2026-vispa</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.2173/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>49402</start>
            <end>49431</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T VISPA: Pluralistic Alignment via Automatic Value Selection and Activation
%A Zheng, Shenyan
%A Zhong, Jiayou
%A Shetty, Anudeex
%A Ji, Heng
%A Nakov, Preslav
%A Naseem, Usman
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F zheng-etal-2026-vispa
%X As large language models are increasingly deployed in high-stakes domains, it is essential that their outputs reflect not average human preference, but diverse perspectives. Achieving such pluralism remains challenging: existing pluralistic alignment methods rely on prompt-level interventions, represent only a limited set of values, or are compute-heavy, lacking explicit value control and representation. Furthermore, existing steering methods require manual value selection and cannot dynamically adapt to scenario-dependent value requirements, limiting their scalability and generality. To address these limitations, we introduce VISPA, a training-free pluralistic alignment framework that performs automatic, context-driven value selection and modular value-steered generation. VISPA identifies scenario-relevant values dynamically and applies activation steering to produce value-conditioned responses representing distinct perspectives. These responses enable support for three pluralistic alignment modes: Overton, Steerable, and Distributional. VISPA supports the addition of new values without finetuning and is compatible with diverse steering instantiations and steering models. Across extensive experiments spanning multiple models and evaluation settings, VISPA demonstrates strong performance across all pluralistic alignment modes in healthcare and beyond.
%U https://aclanthology.org/2026.emnlp-main.2173/
%P 49402-49431

Markdown (Informal)

[VISPA: Pluralistic Alignment via Automatic Value Selection and Activation](https://aclanthology.org/2026.emnlp-main.2173/) (Zheng et al., EMNLP 2026)

• VISPA: Pluralistic Alignment via Automatic Value Selection and Activation (Zheng et al., EMNLP 2026)

ACL

• Shenyan Zheng, Jiayou Zhong, Anudeex Shetty, Heng Ji, Preslav Nakov, and Usman Naseem. 2026. VISPA: Pluralistic Alignment via Automatic Value Selection and Activation. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 49402–49431, Budapest, Hungary. Association for Computational Linguistics.