<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chi-etal-2025-thoughtsculpt">
    <titleInfo>
        <title>ThoughtSculpt: Reasoning with Intermediate Revision and Search</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yizhou</namePart>
        <namePart type="family">Chi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Kevin</namePart>
        <namePart type="family">Yang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dan</namePart>
        <namePart type="family">Klein</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-04</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: NAACL 2025</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Luis</namePart>
            <namePart type="family">Chiruzzo</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Alan</namePart>
            <namePart type="family">Ritter</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Lu</namePart>
            <namePart type="family">Wang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Albuquerque, New Mexico</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-195-7</identifier>
    </relatedItem>
    <abstract>We present THOUGHTSCULPT, a general reasoning and search method for tasks with outputs that can be decomposed into components. THOUGHTSCULPT explores a search tree of potential solutions using Monte Carlo Tree Search (MCTS), building solutions one action at a time and evaluating according to any domain-specific heuristic, which in practice is often simply an LLM evaluator. Critically, our action space includes revision actions: THOUGHTSCULPT may choose to revise part of its previous output rather than continuing to build the rest of its output. Empirically, THOUGHTSCULPT outperforms state-of-the-art reasoning methods across three challenging tasks: Story Outline Improvement (up to +30% interestingness), Mini-Crosswords Solving (up to +16% word success rate), and Constrained Generation (up to +10% concept coverage).</abstract>
    <identifier type="citekey">chi-etal-2025-thoughtsculpt</identifier>
    <identifier type="doi">10.18653/v1/2025.findings-naacl.428</identifier>
    <location>
        <url>https://aclanthology.org/2025.findings-naacl.428/</url>
    </location>
    <part>
        <date>2025-04</date>
        <extent unit="page">
            <start>7685</start>
            <end>7711</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T ThoughtSculpt: Reasoning with Intermediate Revision and Search
%A Chi, Yizhou
%A Yang, Kevin
%A Klein, Dan
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Findings of the Association for Computational Linguistics: NAACL 2025
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-195-7
%F chi-etal-2025-thoughtsculpt
%X We present THOUGHTSCULPT, a general reasoning and search method for tasks with outputs that can be decomposed into components. THOUGHTSCULPT explores a search tree of potential solutions using Monte Carlo Tree Search (MCTS), building solutions one action at a time and evaluating according to any domain-specific heuristic, which in practice is often simply an LLM evaluator. Critically, our action space includes revision actions: THOUGHTSCULPT may choose to revise part of its previous output rather than continuing to build the rest of its output. Empirically, THOUGHTSCULPT outperforms state-of-the-art reasoning methods across three challenging tasks: Story Outline Improvement (up to +30% interestingness), Mini-Crosswords Solving (up to +16% word success rate), and Constrained Generation (up to +10% concept coverage).
%R 10.18653/v1/2025.findings-naacl.428
%U https://aclanthology.org/2025.findings-naacl.428/
%U https://doi.org/10.18653/v1/2025.findings-naacl.428
%P 7685-7711

Markdown (Informal)

[ThoughtSculpt: Reasoning with Intermediate Revision and Search](https://aclanthology.org/2025.findings-naacl.428/) (Chi et al., Findings 2025)

• ThoughtSculpt: Reasoning with Intermediate Revision and Search (Chi et al., Findings 2025)

ACL

• Yizhou Chi, Kevin Yang, and Dan Klein. 2025. ThoughtSculpt: Reasoning with Intermediate Revision and Search. In Findings of the Association for Computational Linguistics: NAACL 2025, pages 7685–7711, Albuquerque, New Mexico. Association for Computational Linguistics.