<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="muti-etal-2026-multitudes">
    <titleInfo>
        <title>I Am Multitudes: A Survey and Roadmap for Rethinking Identity in NLP</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Arianna</namePart>
        <namePart type="family">Muti</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Elisa</namePart>
        <namePart type="family">Bassignana</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Debora</namePart>
        <namePart type="family">Nozza</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>Natural Language Processing (NLP) systems engage with identity across datasets, models, tasks, and evaluation frameworks, yet they often do so by reducing people to fixed categories (e.g., race, gender, age). This reduction is particularly consequential in identity-sensitive applications: for example, it affects how systems classify harm or generate tailored responses to user needs. But individuals contain multitudes: they are not a homogeneous group, but are situated, context-dependent selves whose differences exceed fixed categories. To examine how identity is addressed in NLP, we survey ACL Anthology papers that mention “identity”. Our study investigates how the term is operationalized, where it enters the NLP pipeline, which tasks incorporate identity-related information, and whether its use is explicit or implicit. Our analysis shows that identity is most often operationalized through explicit sociodemographic categories, especially those referring to protected groups by law. Moreover, it is primarily used during the evaluation stage to audit systems for bias and harm. We synthesize these patterns into a taxonomy of identity operationalizations in NLP. Finally, we propose a roadmap to treat identity as relational, contextual, intersectional, and linguistically indexed, with the broader scope of achieving socially aware NLP.</abstract>
    <identifier type="citekey">muti-etal-2026-multitudes</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.1314/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>29912</start>
            <end>29931</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T I Am Multitudes: A Survey and Roadmap for Rethinking Identity in NLP
%A Muti, Arianna
%A Bassignana, Elisa
%A Nozza, Debora
%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 muti-etal-2026-multitudes
%X Natural Language Processing (NLP) systems engage with identity across datasets, models, tasks, and evaluation frameworks, yet they often do so by reducing people to fixed categories (e.g., race, gender, age). This reduction is particularly consequential in identity-sensitive applications: for example, it affects how systems classify harm or generate tailored responses to user needs. But individuals contain multitudes: they are not a homogeneous group, but are situated, context-dependent selves whose differences exceed fixed categories. To examine how identity is addressed in NLP, we survey ACL Anthology papers that mention “identity”. Our study investigates how the term is operationalized, where it enters the NLP pipeline, which tasks incorporate identity-related information, and whether its use is explicit or implicit. Our analysis shows that identity is most often operationalized through explicit sociodemographic categories, especially those referring to protected groups by law. Moreover, it is primarily used during the evaluation stage to audit systems for bias and harm. We synthesize these patterns into a taxonomy of identity operationalizations in NLP. Finally, we propose a roadmap to treat identity as relational, contextual, intersectional, and linguistically indexed, with the broader scope of achieving socially aware NLP.
%U https://aclanthology.org/2026.emnlp-main.1314/
%P 29912-29931

Markdown (Informal)

[I Am Multitudes: A Survey and Roadmap for Rethinking Identity in NLP](https://aclanthology.org/2026.emnlp-main.1314/) (Muti et al., EMNLP 2026)

• I Am Multitudes: A Survey and Roadmap for Rethinking Identity in NLP (Muti et al., EMNLP 2026)

ACL

• Arianna Muti, Elisa Bassignana, and Debora Nozza. 2026. I Am Multitudes: A Survey and Roadmap for Rethinking Identity in NLP. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 29912–29931, Budapest, Hungary. Association for Computational Linguistics.