<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sun-etal-2026-backfire">
    <titleInfo>
        <title>Backfire-R1: Identifying and Mitigating Persuasion Vulnerabilities in LLM Agents</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Jingwei</namePart>
        <namePart type="family">Sun</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yi</namePart>
        <namePart type="family">Yu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jingjing</namePart>
        <namePart type="family">Qu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hanxi</namePart>
        <namePart type="family">Zhu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jia</namePart>
        <namePart type="family">Xu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jing</namePart>
        <namePart type="family">Shao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-462-0</identifier>
    </relatedItem>
    <abstract>With the rapid development of Large Language Models (LLMs), research related to LLM-based persuasion has become a hotspot in the field. However, existing studies mostly focus on enhancing LLM agents’ persuasive capabilities, while generally neglecting the issue of their vulnerability to persuasion. In this paper, we analyse the root causes of LLM agents’ vulnerability to persuasion and find that it mainly stems from two contradictions and limitations. First, the purpose of catering to human preferences in pre-training phase conflicts with the need to resist manipulation in persuasion scenarios. Second, LLMs are often restricted to play a helpful assistant during training, and the singularity of this personality setting makes it unusually easy to develop targeted persuasion techniques for them. To eliminate LLMs’ vulnerability to persuasion, we propose the Backfire-R1 framework, which helps LLMs build persuasion resistance and better align with user objective by enabling them to learn the reasoning logic of humans with different personalities in persuasion scenarios. Comprehensive experimental results show that, the proposed method significantly improves the persuasion robustness of LLM agents, reducing the opinion shift score by 62.36% while not impairing the model’s general capabilities. Code is now available at: https://github.com/super-jw/Backfire-R1.</abstract>
    <identifier type="citekey">sun-etal-2026-backfire</identifier>
    <identifier type="doi">10.18653/2026.findings-emnlp.601</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.601/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>12674</start>
            <end>12701</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Backfire-R1: Identifying and Mitigating Persuasion Vulnerabilities in LLM Agents
%A Sun, Jingwei
%A Yu, Yi
%A Qu, Jingjing
%A Zhu, Hanxi
%A Xu, Jia
%A Shao, Jing
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F sun-etal-2026-backfire
%X With the rapid development of Large Language Models (LLMs), research related to LLM-based persuasion has become a hotspot in the field. However, existing studies mostly focus on enhancing LLM agents’ persuasive capabilities, while generally neglecting the issue of their vulnerability to persuasion. In this paper, we analyse the root causes of LLM agents’ vulnerability to persuasion and find that it mainly stems from two contradictions and limitations. First, the purpose of catering to human preferences in pre-training phase conflicts with the need to resist manipulation in persuasion scenarios. Second, LLMs are often restricted to play a helpful assistant during training, and the singularity of this personality setting makes it unusually easy to develop targeted persuasion techniques for them. To eliminate LLMs’ vulnerability to persuasion, we propose the Backfire-R1 framework, which helps LLMs build persuasion resistance and better align with user objective by enabling them to learn the reasoning logic of humans with different personalities in persuasion scenarios. Comprehensive experimental results show that, the proposed method significantly improves the persuasion robustness of LLM agents, reducing the opinion shift score by 62.36% while not impairing the model’s general capabilities. Code is now available at: https://github.com/super-jw/Backfire-R1.
%R 10.18653/2026.findings-emnlp.601
%U https://aclanthology.org/2026.findings-emnlp.601/
%U https://doi.org/10.18653/2026.findings-emnlp.601
%P 12674-12701

Markdown (Informal)

[Backfire-R1: Identifying and Mitigating Persuasion Vulnerabilities in LLM Agents](https://aclanthology.org/2026.findings-emnlp.601/) (Sun et al., Findings 2026)

• Backfire-R1: Identifying and Mitigating Persuasion Vulnerabilities in LLM Agents (Sun et al., Findings 2026)

ACL

• Jingwei Sun, Yi Yu, Jingjing Qu, Hanxi Zhu, Jia Xu, and Jing Shao. 2026. Backfire-R1: Identifying and Mitigating Persuasion Vulnerabilities in LLM Agents. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 12674–12701, Budapest, Hungary. Association for Computational Linguistics.