<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2026-ratf">
    <titleInfo>
        <title>RATF: Robust Adaptive Tri-Encoder Framework for Fake News Detection against Style-Confusion Attacks from LLMs</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Haoda</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zhenhua</namePart>
        <namePart type="family">Tan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ziwei</namePart>
        <namePart type="family">Cheng</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>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</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">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>The rise of Large Language Models (LLMs) enables a new challenge for fake news detection: style-confusion attacks, where attackers rewrite news to mimic trusted styles, challenging detectors that rely on stylistic features. Current defenses like style-invariance training ignore stylistic features, but this is suboptimal for style-confusion attack datasets as it discards valuable cues in clean (non-attacked) articles. To defend against style-confusion attacks, we propose the Robust Adaptive Tri-Encoder Framework (RATF). Our framework uses an LLM to decouple news into content features and style features, and leverages conditional probability curvature to compute a “rewritten score” estimating the likelihood that the article has been rewritten by an LLM. RATF injects this score into a score-conditioned cross-attention fusion module and a learnable parametric gating module. The former uses the content representation as a reliable query to retrieve relevant semantic and stylistic evidence, while the latter continuously calibrates the contributions of original semantics, content features, and style features. We simulate these attacks by creating an equal mixture of original articles from the PolitiFact and GossipCop datasets and their four corresponding style-rewritten versions generated by an LLM. Our experiments show RATF achieves state-of-the-art performance with superior robustness against style-confusion attacks, while also proving highly adaptable across various backbones.</abstract>
    <identifier type="citekey">zhang-etal-2026-ratf</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.1303/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>29683</start>
            <end>29694</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T RATF: Robust Adaptive Tri-Encoder Framework for Fake News Detection against Style-Confusion Attacks from LLMs
%A Zhang, Haoda
%A Tan, Zhenhua
%A Cheng, Ziwei
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F zhang-etal-2026-ratf
%X The rise of Large Language Models (LLMs) enables a new challenge for fake news detection: style-confusion attacks, where attackers rewrite news to mimic trusted styles, challenging detectors that rely on stylistic features. Current defenses like style-invariance training ignore stylistic features, but this is suboptimal for style-confusion attack datasets as it discards valuable cues in clean (non-attacked) articles. To defend against style-confusion attacks, we propose the Robust Adaptive Tri-Encoder Framework (RATF). Our framework uses an LLM to decouple news into content features and style features, and leverages conditional probability curvature to compute a “rewritten score” estimating the likelihood that the article has been rewritten by an LLM. RATF injects this score into a score-conditioned cross-attention fusion module and a learnable parametric gating module. The former uses the content representation as a reliable query to retrieve relevant semantic and stylistic evidence, while the latter continuously calibrates the contributions of original semantics, content features, and style features. We simulate these attacks by creating an equal mixture of original articles from the PolitiFact and GossipCop datasets and their four corresponding style-rewritten versions generated by an LLM. Our experiments show RATF achieves state-of-the-art performance with superior robustness against style-confusion attacks, while also proving highly adaptable across various backbones.
%U https://aclanthology.org/2026.emnlp-main.1303/
%P 29683-29694

Markdown (Informal)

[RATF: Robust Adaptive Tri-Encoder Framework for Fake News Detection against Style-Confusion Attacks from LLMs](https://aclanthology.org/2026.emnlp-main.1303/) (Zhang et al., EMNLP 2026)

• RATF: Robust Adaptive Tri-Encoder Framework for Fake News Detection against Style-Confusion Attacks from LLMs (Zhang et al., EMNLP 2026)

ACL

• Haoda Zhang, Zhenhua Tan, and Ziwei Cheng. 2026. RATF: Robust Adaptive Tri-Encoder Framework for Fake News Detection against Style-Confusion Attacks from LLMs. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 29683–29694, Budapest, Hungary. Association for Computational Linguistics.