<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhong-etal-2025-dyve">
    <titleInfo>
        <title>Dyve: Thinking Fast and Slow for Dynamic Process Verification</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Jianyuan</namePart>
        <namePart type="family">Zhong</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zeju</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zhijian</namePart>
        <namePart type="family">Xu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xiangyu</namePart>
        <namePart type="family">Wen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Qiang</namePart>
        <namePart type="family">Xu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Christos</namePart>
            <namePart type="family">Christodoulopoulos</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tanmoy</namePart>
            <namePart type="family">Chakraborty</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Carolyn</namePart>
            <namePart type="family">Rose</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Violet</namePart>
            <namePart type="family">Peng</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Suzhou, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-332-6</identifier>
    </relatedItem>
    <abstract>Large Language Models have advanced significantly in complex reasoning, often leveraging external reward model to improve the reliability of their multi-step processes. However, existing process verification methods struggle with reliably assessing incomplete reasoning traces and are limited by the cost of high-quality human annotations or the inherent noise in automatically generated labels. Therefore, we present Dyve, a dynamic process verifier that enhances reasoning error detection in large language models by integrating fast and slow thinking, inspired by Kahneman’s Systems Theory. Dyve adaptively applies immediate token-level confirmation (System 1) for straightforward steps and comprehensive analysis (System 2) for complex ones. Unlike traditional verifiers that only evaluate final outputs, Dyve employs a step-wise consensus-filtered supervision strategy, leveraging Monte Carlo estimation, LLM-as-a-Judge, and specialized reasoning models to extract high-quality training signals from noisy rollouts. Experimental results on ProcessBench and the MATH dataset confirm that Dyve significantly outperforms existing process-based verifiers and boosts performance in Best-of-N settings while maintaining computational efficiency by strategically allocating verification resources.</abstract>
    <identifier type="citekey">zhong-etal-2025-dyve</identifier>
    <location>
        <url>https://aclanthology.org/2025.emnlp-main.1136/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>22331</start>
            <end>22344</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Dyve: Thinking Fast and Slow for Dynamic Process Verification
%A Zhong, Jianyuan
%A Li, Zeju
%A Xu, Zhijian
%A Wen, Xiangyu
%A Xu, Qiang
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-332-6
%F zhong-etal-2025-dyve
%X Large Language Models have advanced significantly in complex reasoning, often leveraging external reward model to improve the reliability of their multi-step processes. However, existing process verification methods struggle with reliably assessing incomplete reasoning traces and are limited by the cost of high-quality human annotations or the inherent noise in automatically generated labels. Therefore, we present Dyve, a dynamic process verifier that enhances reasoning error detection in large language models by integrating fast and slow thinking, inspired by Kahneman’s Systems Theory. Dyve adaptively applies immediate token-level confirmation (System 1) for straightforward steps and comprehensive analysis (System 2) for complex ones. Unlike traditional verifiers that only evaluate final outputs, Dyve employs a step-wise consensus-filtered supervision strategy, leveraging Monte Carlo estimation, LLM-as-a-Judge, and specialized reasoning models to extract high-quality training signals from noisy rollouts. Experimental results on ProcessBench and the MATH dataset confirm that Dyve significantly outperforms existing process-based verifiers and boosts performance in Best-of-N settings while maintaining computational efficiency by strategically allocating verification resources.
%U https://aclanthology.org/2025.emnlp-main.1136/
%P 22331-22344

Markdown (Informal)

[Dyve: Thinking Fast and Slow for Dynamic Process Verification](https://aclanthology.org/2025.emnlp-main.1136/) (Zhong et al., EMNLP 2025)

• Dyve: Thinking Fast and Slow for Dynamic Process Verification (Zhong et al., EMNLP 2025)

ACL

• Jianyuan Zhong, Zeju Li, Zhijian Xu, Xiangyu Wen, and Qiang Xu. 2025. Dyve: Thinking Fast and Slow for Dynamic Process Verification. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 22331–22344, Suzhou, China. Association for Computational Linguistics.