<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="anh-etal-2026-circuit">
    <titleInfo>
        <title>A Circuit for Plural Reference: How LLMs Represent and Retrieve Singular and Plural Entities</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Dang</namePart>
        <namePart type="given">Thi</namePart>
        <namePart type="given">Thao</namePart>
        <namePart type="family">Anh</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Rick</namePart>
        <namePart type="family">Nouwen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Massimo</namePart>
        <namePart type="family">Poesio</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>Coreference resolution is an important task in contextual reasoning. In this paper, we investigate the mechanism for representing and retrieving singular and plural entities for plural reference. We use a combination of mechanistic interpretability and attention pattern analysis to study the process in which LLMs predict a pronoun to refer back to previously mentioned entities. Using a range of causal intervention techniques, we find a set of attention heads that are responsible for (1) representing coreference information in the input, (2) identifying entities that form a plural reference, (3) transferring the information to the component that is responsible for selecting the antecedents and predicting the pronoun. We also find that LLMs align with humans in preference for plural pronoun. Specifically, entities in a plural construction are more likely to be referred to as a plural entity if they are ontologically similar and are linked by the conjunction and.</abstract>
    <identifier type="citekey">anh-etal-2026-circuit</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.1759/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>37627</start>
            <end>37644</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T A Circuit for Plural Reference: How LLMs Represent and Retrieve Singular and Plural Entities
%A Anh, Dang Thi Thao
%A Nouwen, Rick
%A Poesio, Massimo
%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 anh-etal-2026-circuit
%X Coreference resolution is an important task in contextual reasoning. In this paper, we investigate the mechanism for representing and retrieving singular and plural entities for plural reference. We use a combination of mechanistic interpretability and attention pattern analysis to study the process in which LLMs predict a pronoun to refer back to previously mentioned entities. Using a range of causal intervention techniques, we find a set of attention heads that are responsible for (1) representing coreference information in the input, (2) identifying entities that form a plural reference, (3) transferring the information to the component that is responsible for selecting the antecedents and predicting the pronoun. We also find that LLMs align with humans in preference for plural pronoun. Specifically, entities in a plural construction are more likely to be referred to as a plural entity if they are ontologically similar and are linked by the conjunction and.
%U https://aclanthology.org/2026.findings-emnlp.1759/
%P 37627-37644

Markdown (Informal)

[A Circuit for Plural Reference: How LLMs Represent and Retrieve Singular and Plural Entities](https://aclanthology.org/2026.findings-emnlp.1759/) (Anh et al., Findings 2026)

• A Circuit for Plural Reference: How LLMs Represent and Retrieve Singular and Plural Entities (Anh et al., Findings 2026)

ACL

• Dang Thi Thao Anh, Rick Nouwen, and Massimo Poesio. 2026. A Circuit for Plural Reference: How LLMs Represent and Retrieve Singular and Plural Entities. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 37627–37644, Budapest, Hungary. Association for Computational Linguistics.