<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2025-adaptive">
    <titleInfo>
        <title>Adaptive Coopetition: Leveraging Coarse Verifier Signals for Resilient Multi-Agent LLM Reasoning</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Wendy</namePart>
        <namePart type="given">Yaqiao</namePart>
        <namePart type="family">Liu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Rui</namePart>
        <namePart type="given">Jerry</namePart>
        <namePart type="family">Huang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Anastasia</namePart>
        <namePart type="family">Miin</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lei</namePart>
        <namePart type="family">Ding</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-12</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>The 14th International Joint Conference on Natural Language Processing and The 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Santosh</namePart>
            <namePart type="family">T.y.s.s</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Shuichiro</namePart>
            <namePart type="family">Shimizu</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yifan</namePart>
            <namePart type="family">Gong</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Mumbai, India</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-304-3</identifier>
    </relatedItem>
    <abstract>Inference-time computation is a critical yet challenging paradigm for enhancing the reasoning performance of large language models (LLMs). While existing strategies improve reasoning stability and consistency, they suffer from notable limitations: self-correction often reinforces the model’s initial biases, and Multi-Agent Collaboration (MAC) often fails due to the lack of efficient coordination mechanisms, leading to collective errors. Although high-performing verifiers can detect reasoning errors, making them reliable requires substantial training. To address these challenges, we introduce a novel inference-time framework - **Adaptive Coopetition (AdCo)** - in which LLM agents utilize **an adaptive, UCB-based ‘coopetition’ mechanism**. At each round, agents leverage coarse verifier signals to determine whether to collaborate or compete, further iteratively refining their reasoning based on peer feedback. Without relying on high-performance verifiers, our adaptive strategy achieves significant performance gains on mathematical reasoning benchmarks, yielding **a 20% relative improvement** over baselines on the more challenging dataset. Our approach remains robust and consistent in terms of accuracy under different sample sizes and configurations. This adaptive, signal-guided ‘coopetition’ framework enhances reasoning robustness by leveraging bothmodel knowledge diversity and reasoning trace measure, while also promoting uncertainty-driven exploration, especially when participants have comparable capabilities. From this perspective, our work offers a fresh lens on inference-time computation and paves the way for more resilient multi-agent LLM systems.</abstract>
    <identifier type="citekey">liu-etal-2025-adaptive</identifier>
    <location>
        <url>https://aclanthology.org/2025.ijcnlp-srw.13/</url>
    </location>
    <part>
        <date>2025-12</date>
        <extent unit="page">
            <start>145</start>
            <end>155</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Adaptive Coopetition: Leveraging Coarse Verifier Signals for Resilient Multi-Agent LLM Reasoning
%A Liu, Wendy Yaqiao
%A Huang, Rui Jerry
%A Miin, Anastasia
%A Ding, Lei
%Y T.y.s.s, Santosh
%Y Shimizu, Shuichiro
%Y Gong, Yifan
%S The 14th International Joint Conference on Natural Language Processing and The 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics
%D 2025
%8 December
%I Association for Computational Linguistics
%C Mumbai, India
%@ 979-8-89176-304-3
%F liu-etal-2025-adaptive
%X Inference-time computation is a critical yet challenging paradigm for enhancing the reasoning performance of large language models (LLMs). While existing strategies improve reasoning stability and consistency, they suffer from notable limitations: self-correction often reinforces the model’s initial biases, and Multi-Agent Collaboration (MAC) often fails due to the lack of efficient coordination mechanisms, leading to collective errors. Although high-performing verifiers can detect reasoning errors, making them reliable requires substantial training. To address these challenges, we introduce a novel inference-time framework - **Adaptive Coopetition (AdCo)** - in which LLM agents utilize **an adaptive, UCB-based ‘coopetition’ mechanism**. At each round, agents leverage coarse verifier signals to determine whether to collaborate or compete, further iteratively refining their reasoning based on peer feedback. Without relying on high-performance verifiers, our adaptive strategy achieves significant performance gains on mathematical reasoning benchmarks, yielding **a 20% relative improvement** over baselines on the more challenging dataset. Our approach remains robust and consistent in terms of accuracy under different sample sizes and configurations. This adaptive, signal-guided ‘coopetition’ framework enhances reasoning robustness by leveraging bothmodel knowledge diversity and reasoning trace measure, while also promoting uncertainty-driven exploration, especially when participants have comparable capabilities. From this perspective, our work offers a fresh lens on inference-time computation and paves the way for more resilient multi-agent LLM systems.
%U https://aclanthology.org/2025.ijcnlp-srw.13/
%P 145-155

Markdown (Informal)

[Adaptive Coopetition: Leveraging Coarse Verifier Signals for Resilient Multi-Agent LLM Reasoning](https://aclanthology.org/2025.ijcnlp-srw.13/) (Liu et al., IJCNLP 2025)

• Adaptive Coopetition: Leveraging Coarse Verifier Signals for Resilient Multi-Agent LLM Reasoning (Liu et al., IJCNLP 2025)

ACL

• Wendy Yaqiao Liu, Rui Jerry Huang, Anastasia Miin, and Lei Ding. 2025. Adaptive Coopetition: Leveraging Coarse Verifier Signals for Resilient Multi-Agent LLM Reasoning. In The 14th International Joint Conference on Natural Language Processing and The 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics, pages 145–155, Mumbai, India. Association for Computational Linguistics.