<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hoyle-etal-2025-proxann">
    <titleInfo>
        <title>ProxAnn: Use-Oriented Evaluations of Topic Models and Document Clustering</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Alexander</namePart>
        <namePart type="family">Hoyle</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lorena</namePart>
        <namePart type="family">Calvo-Bartolomé</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jordan</namePart>
        <namePart type="family">Boyd-Graber</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Philip</namePart>
        <namePart type="family">Resnik</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>Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</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">979-8-89176-251-0</identifier>
    </relatedItem>
    <abstract>Topic models and document-clustering evaluations either use automated metrics that align poorly with human preferences, or require expert labels that are intractable to scale. We design a scalable human evaluation protocol and a corresponding automated approximation that reflect practitioners’ real-world usage of models. Annotators—or an LLM-based proxy—review text items assigned to a topic or cluster, infer a category for the group, then apply that category to other documents. Using this protocol, we collect extensive crowdworker annotations of outputs from a diverse set of topic models on two datasets. We then use these annotations to validate automated proxies, finding that the best LLM proxy is statistically indistinguishable from a human annotator and can therefore serve as a reasonable substitute in automated evaluations.</abstract>
    <identifier type="citekey">hoyle-etal-2025-proxann</identifier>
    <identifier type="doi">10.18653/v1/2025.acl-long.772</identifier>
    <location>
        <url>https://aclanthology.org/2025.acl-long.772/</url>
    </location>
    <part>
        <date>2025-07</date>
        <extent unit="page">
            <start>15872</start>
            <end>15897</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T ProxAnn: Use-Oriented Evaluations of Topic Models and Document Clustering
%A Hoyle, Alexander
%A Calvo-Bartolomé, Lorena
%A Boyd-Graber, Jordan
%A Resnik, Philip
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-251-0
%F hoyle-etal-2025-proxann
%X Topic models and document-clustering evaluations either use automated metrics that align poorly with human preferences, or require expert labels that are intractable to scale. We design a scalable human evaluation protocol and a corresponding automated approximation that reflect practitioners’ real-world usage of models. Annotators—or an LLM-based proxy—review text items assigned to a topic or cluster, infer a category for the group, then apply that category to other documents. Using this protocol, we collect extensive crowdworker annotations of outputs from a diverse set of topic models on two datasets. We then use these annotations to validate automated proxies, finding that the best LLM proxy is statistically indistinguishable from a human annotator and can therefore serve as a reasonable substitute in automated evaluations.
%R 10.18653/v1/2025.acl-long.772
%U https://aclanthology.org/2025.acl-long.772/
%U https://doi.org/10.18653/v1/2025.acl-long.772
%P 15872-15897

Markdown (Informal)

[ProxAnn: Use-Oriented Evaluations of Topic Models and Document Clustering](https://aclanthology.org/2025.acl-long.772/) (Hoyle et al., ACL 2025)

• ProxAnn: Use-Oriented Evaluations of Topic Models and Document Clustering (Hoyle et al., ACL 2025)

ACL

• Alexander Hoyle, Lorena Calvo-Bartolomé, Jordan Boyd-Graber, and Philip Resnik. 2025. ProxAnn: Use-Oriented Evaluations of Topic Models and Document Clustering. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 15872–15897, Vienna, Austria. Association for Computational Linguistics.