<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ying-etal-2025-understanding">
    <titleInfo>
        <title>Understanding Epistemic Language with a Language-augmented Bayesian Theory of Mind</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Lance</namePart>
        <namePart type="family">Ying</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tan</namePart>
        <namePart type="family">Zhi-Xuan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lionel</namePart>
        <namePart type="family">Wong</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Vikash</namePart>
        <namePart type="family">Mansinghka</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Joshua</namePart>
        <namePart type="given">B</namePart>
        <namePart type="family">Tenenbaum</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025</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>How do people understand and evaluate claims about others’ beliefs, even though these beliefs cannot be directly observed? In this paper, we introduce a cognitive model of epistemic language interpretation, grounded in Bayesian inferences about other agents’ goals, beliefs, and intentions: a language-augmented Bayesian theory-of-mind (LaBToM). By translating natural language into an epistemic “language-of-thought” with grammar-constrained LLM decoding, then evaluating these translations against the inferences produced by inverting a generative model of rational action and perception, LaBToM captures graded plausibility judgments of epistemic claims. We validate our model in an experiment where participants watch an agent navigate a maze to find keys hidden in boxes needed to reach their goal, then rate sentences about the agent’s beliefs. In contrast with multimodal LLMs (GPT-4o, Gemini Pro) and ablated models, our model correlates highly with human judgments for a wide range of expressions, including modal language, uncertainty expressions, knowledge claims, likelihood comparisons, and attributions of false belief.</abstract>
    <identifier type="citekey">ying-etal-2025-understanding</identifier>
    <identifier type="doi">10.1162/tacl_a_00752</identifier>
    <location>
        <url>https://aclanthology.org/2025.tacl-1.30/</url>
    </location>
    <part>
        <date>2025</date>
        <detail type="volume"><number>13</number></detail>
        <extent unit="page">
            <start>613</start>
            <end>637</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T Understanding Epistemic Language with a Language-augmented Bayesian Theory of Mind
%A Ying, Lance
%A Zhi-Xuan, Tan
%A Wong, Lionel
%A Mansinghka, Vikash
%A Tenenbaum, Joshua B.
%J Transactions of the Association for Computational Linguistics
%D 2025
%V 13
%I MIT Press
%C Cambridge, MA
%F ying-etal-2025-understanding
%X How do people understand and evaluate claims about others’ beliefs, even though these beliefs cannot be directly observed? In this paper, we introduce a cognitive model of epistemic language interpretation, grounded in Bayesian inferences about other agents’ goals, beliefs, and intentions: a language-augmented Bayesian theory-of-mind (LaBToM). By translating natural language into an epistemic “language-of-thought” with grammar-constrained LLM decoding, then evaluating these translations against the inferences produced by inverting a generative model of rational action and perception, LaBToM captures graded plausibility judgments of epistemic claims. We validate our model in an experiment where participants watch an agent navigate a maze to find keys hidden in boxes needed to reach their goal, then rate sentences about the agent’s beliefs. In contrast with multimodal LLMs (GPT-4o, Gemini Pro) and ablated models, our model correlates highly with human judgments for a wide range of expressions, including modal language, uncertainty expressions, knowledge claims, likelihood comparisons, and attributions of false belief.
%R 10.1162/tacl_a_00752
%U https://aclanthology.org/2025.tacl-1.30/
%U https://doi.org/10.1162/tacl_a_00752
%P 613-637

Markdown (Informal)

[Understanding Epistemic Language with a Language-augmented Bayesian Theory of Mind](https://aclanthology.org/2025.tacl-1.30/) (Ying et al., TACL 2025)

• Understanding Epistemic Language with a Language-augmented Bayesian Theory of Mind (Ying et al., TACL 2025)

ACL

• Lance Ying, Tan Zhi-Xuan, Lionel Wong, Vikash Mansinghka, and Joshua B. Tenenbaum. 2025. Understanding Epistemic Language with a Language-augmented Bayesian Theory of Mind. Transactions of the Association for Computational Linguistics, 13:613–637.