<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2025-elicit">
    <titleInfo>
        <title>Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Xingxuan</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Weiwen</namePart>
        <namePart type="family">Xu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ruochen</namePart>
        <namePart type="family">Zhao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Fangkai</namePart>
        <namePart type="family">Jiao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Shafiq</namePart>
        <namePart type="family">Joty</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lidong</namePart>
        <namePart type="family">Bing</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>Large language models excel at problem-solving but often struggle with complex reasoning and factual accuracy. While chain-of-thought and retrieval-augmented generation help break down problems and retrieve knowledge, they still falter on challenging tasks like competitive programming due to frequent reasoning errors and irrelevant retrieval. To address this, we introduce Critic-guided planning with Retrieval-augmentation, CR-Planner, a novel framework that leverages fine-tuned critic models to guide both reasoning and retrieval processes through planning. CR-Planner iteratively selects and executes sub-goals, guided by critic models. A sub-goal critic identifies promising sub-goals from reasoning, query generation, and retrieval, while an execution critic evaluates outputs of sub-goal executions. We employ Monte Carlo Tree Search to collect data for critic training, allowing systematic exploration of action sequences and effective navigation toward the final answer. We evaluate CR-Planner on challenging domain-knowledge-intensive and reasoning-heavy tasks, including competitive programming, theorem-driven math reasoning, and complex domain retrieval problems. It significantly outperforms baselines, demonstrating effectiveness in both reasoning and retrieval.</abstract>
    <identifier type="citekey">li-etal-2025-elicit</identifier>
    <identifier type="doi">10.18653/v1/2025.acl-long.1244</identifier>
    <location>
        <url>https://aclanthology.org/2025.acl-long.1244/</url>
    </location>
    <part>
        <date>2025-07</date>
        <extent unit="page">
            <start>25589</start>
            <end>25604</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks
%A Li, Xingxuan
%A Xu, Weiwen
%A Zhao, Ruochen
%A Jiao, Fangkai
%A Joty, Shafiq
%A Bing, Lidong
%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 li-etal-2025-elicit
%X Large language models excel at problem-solving but often struggle with complex reasoning and factual accuracy. While chain-of-thought and retrieval-augmented generation help break down problems and retrieve knowledge, they still falter on challenging tasks like competitive programming due to frequent reasoning errors and irrelevant retrieval. To address this, we introduce Critic-guided planning with Retrieval-augmentation, CR-Planner, a novel framework that leverages fine-tuned critic models to guide both reasoning and retrieval processes through planning. CR-Planner iteratively selects and executes sub-goals, guided by critic models. A sub-goal critic identifies promising sub-goals from reasoning, query generation, and retrieval, while an execution critic evaluates outputs of sub-goal executions. We employ Monte Carlo Tree Search to collect data for critic training, allowing systematic exploration of action sequences and effective navigation toward the final answer. We evaluate CR-Planner on challenging domain-knowledge-intensive and reasoning-heavy tasks, including competitive programming, theorem-driven math reasoning, and complex domain retrieval problems. It significantly outperforms baselines, demonstrating effectiveness in both reasoning and retrieval.
%R 10.18653/v1/2025.acl-long.1244
%U https://aclanthology.org/2025.acl-long.1244/
%U https://doi.org/10.18653/v1/2025.acl-long.1244
%P 25589-25604

Markdown (Informal)

[Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks](https://aclanthology.org/2025.acl-long.1244/) (Li et al., ACL 2025)

• Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks (Li et al., ACL 2025)

ACL

• Xingxuan Li, Weiwen Xu, Ruochen Zhao, Fangkai Jiao, Shafiq Joty, and Lidong Bing. 2025. Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 25589–25604, Vienna, Austria. Association for Computational Linguistics.