<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dutta-chowdhury-etal-2026-formatting">
    <titleInfo>
        <title>Formatting and Surface Feature Confounds in AI Text Detection: A Clever Hans Effect</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Koel</namePart>
        <namePart type="family">Dutta Chowdhury</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Cristina</namePart>
        <namePart type="family">España-Bonet</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Josef</namePart>
        <namePart type="family">Van Genabith</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>AI-text detectors are typically evaluated on benchmarks that pair human-written text with text generated by language models. However, differences in how these two classes are constructed can introduce surface-form cues that make the benchmark easier to solve. We show that surface form alone strongly separates human from machine-generated text across three detection benchmarks, but the learned surface patterns vary substantially across benchmarks. Detector scores are also substantially predictable from surface form, although detectors retain signal beyond it, and controlled reformatting reveals partial and detector-specific causal effects. Our findings show that benchmark performance alone does not establish that a detector has identified a stable and transferable signal of machine generation.</abstract>
    <identifier type="citekey">dutta-chowdhury-etal-2026-formatting</identifier>
    <identifier type="doi">10.18653/2026.findings-emnlp.1051</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.1051/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>22266</start>
            <end>22277</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Formatting and Surface Feature Confounds in AI Text Detection: A Clever Hans Effect
%A Dutta Chowdhury, Koel
%A España-Bonet, Cristina
%A Van Genabith, Josef
%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 dutta-chowdhury-etal-2026-formatting
%X AI-text detectors are typically evaluated on benchmarks that pair human-written text with text generated by language models. However, differences in how these two classes are constructed can introduce surface-form cues that make the benchmark easier to solve. We show that surface form alone strongly separates human from machine-generated text across three detection benchmarks, but the learned surface patterns vary substantially across benchmarks. Detector scores are also substantially predictable from surface form, although detectors retain signal beyond it, and controlled reformatting reveals partial and detector-specific causal effects. Our findings show that benchmark performance alone does not establish that a detector has identified a stable and transferable signal of machine generation.
%R 10.18653/2026.findings-emnlp.1051
%U https://aclanthology.org/2026.findings-emnlp.1051/
%U https://doi.org/10.18653/2026.findings-emnlp.1051
%P 22266-22277

Markdown (Informal)

[Formatting and Surface Feature Confounds in AI Text Detection: A Clever Hans Effect](https://aclanthology.org/2026.findings-emnlp.1051/) (Dutta Chowdhury et al., Findings 2026)

• Formatting and Surface Feature Confounds in AI Text Detection: A Clever Hans Effect (Dutta Chowdhury et al., Findings 2026)

ACL

• Koel Dutta Chowdhury, Cristina España-Bonet, and Josef Van Genabith. 2026. Formatting and Surface Feature Confounds in AI Text Detection: A Clever Hans Effect. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 22266–22277, Budapest, Hungary. Association for Computational Linguistics.