<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="stevenson-etal-2026-large">
    <titleInfo>
        <title>Can Large Language Models Generalize Analogy Solving Like Children Can?</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Claire</namePart>
        <namePart type="given">E</namePart>
        <namePart type="family">Stevenson</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Alexandra</namePart>
        <namePart type="family">Pafford</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Han</namePart>
        <namePart type="given">L</namePart>
        <namePart type="given">J</namePart>
        <namePart type="family">van der Maas</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Melanie</namePart>
        <namePart type="family">Mitchell</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <genre authority="bibutilsgt">journal article</genre>
    <relatedItem type="host">
        <titleInfo>
            <title>Transactions of the Association for Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <issuance>continuing</issuance>
            <publisher>MIT Press</publisher>
            <place>
                <placeTerm type="text">Cambridge, MA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">periodical</genre>
        <genre authority="bibutilsgt">academic journal</genre>
    </relatedItem>
    <abstract>In people, the ability to solve analogies such as “body: feet:: table: ?” emerges in childhood, and appears to transfer easily to other domains, such as the visual domain “(: ) :: \ensuremath&lt; : ?”. Recent research shows that large language models (LLMs) can solve various forms of analogies. However, can LLMs generalize analogy solving to other domains like people can? To investigate this, we had children, adults, and LLMs solve a series of letter-string analogies (e.g., a b : a c :: j k : ?) in the Latin alphabet, in a near transfer domain (Greek alphabet), and a far transfer domain (list of symbols). Children and adults easily generalized their knowledge to unfamiliar domains, whereas LLMs did not. This key difference between human and AI performance is evidence that these LLMs still struggle with robust human-like analogical transfer.</abstract>
    <identifier type="citekey">stevenson-etal-2026-large</identifier>
    <identifier type="doi">10.1162/tacl.a.614</identifier>
    <location>
        <url>https://aclanthology.org/2026.tacl-1.28/</url>
    </location>
    <part>
        <date>2026</date>
        <detail type="volume"><number>14</number></detail>
        <extent unit="page">
            <start>612</start>
            <end>626</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T Can Large Language Models Generalize Analogy Solving Like Children Can?
%A Stevenson, Claire E.
%A Pafford, Alexandra
%A van der Maas, Han L. J.
%A Mitchell, Melanie
%J Transactions of the Association for Computational Linguistics
%D 2026
%V 14
%I MIT Press
%C Cambridge, MA
%F stevenson-etal-2026-large
%X In people, the ability to solve analogies such as “body: feet:: table: ?” emerges in childhood, and appears to transfer easily to other domains, such as the visual domain “(: ) :: \ensuremath< : ?”. Recent research shows that large language models (LLMs) can solve various forms of analogies. However, can LLMs generalize analogy solving to other domains like people can? To investigate this, we had children, adults, and LLMs solve a series of letter-string analogies (e.g., a b : a c :: j k : ?) in the Latin alphabet, in a near transfer domain (Greek alphabet), and a far transfer domain (list of symbols). Children and adults easily generalized their knowledge to unfamiliar domains, whereas LLMs did not. This key difference between human and AI performance is evidence that these LLMs still struggle with robust human-like analogical transfer.
%R 10.1162/tacl.a.614
%U https://aclanthology.org/2026.tacl-1.28/
%U https://doi.org/10.1162/tacl.a.614
%P 612-626

Markdown (Informal)

[Can Large Language Models Generalize Analogy Solving Like Children Can?](https://aclanthology.org/2026.tacl-1.28/) (Stevenson et al., TACL 2026)

• Can Large Language Models Generalize Analogy Solving Like Children Can? (Stevenson et al., TACL 2026)

ACL

• Claire E. Stevenson, Alexandra Pafford, Han L. J. van der Maas, and Melanie Mitchell. 2026. Can Large Language Models Generalize Analogy Solving Like Children Can?. Transactions of the Association for Computational Linguistics, 14:612–626.