<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="min-etal-2026-reformulating">
    <titleInfo>
        <title>Reformulating Radiology Report Error Detection as Sentence-Level Vision-Language Learning</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Dabin</namePart>
        <namePart type="family">Min</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Kyungmin</namePart>
        <namePart type="family">Jeon</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dongheon</namePart>
        <namePart type="family">Lee</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Chang</namePart>
        <namePart type="given">Min</namePart>
        <namePart type="family">Park</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>Accurate alignment between chest X-ray (CXR) images and radiology reports is essential for clinical safety, yet automatic radiology report error detection from paired CXRs remains challenging. Prior approaches typically compare the entire report with the image, limiting the model’s ability to learn fine-grained vision-language correspondences at the level of individual findings. We reformulate report error detection as a sentence-level vision-language learning problem by decomposing reports into atomic finding statements, thereby reducing semantic entanglement and supervision noise. We propose SenRED (Sentence-level Report Error Detector), which uses sentence-conditioned patch aggregation to associate each statement with localized visual evidence. To supervise learning, we generate sentence-level synthetic errors reflecting clinically relevant categories—false positives, false negatives, and mislocalization—along with consistent image-sentence pairs. Experiments on MIMIC-CXR and Open-i show that SenRED consistently outperforms task-specific baselines and zero-shot instruction-tuned VLMs in both sentence-level and exam-level evaluations. Radiologist-confirmed evaluations further show that these gains are maintained on clinically reviewed subsets. Comparative experiments support the benefit of sentence-level reformulation, while ablations demonstrate that sentence-conditioned patch aggregation improves fine-grained consistency checking.</abstract>
    <identifier type="citekey">min-etal-2026-reformulating</identifier>
    <identifier type="doi">10.18653/2026.findings-emnlp.2295</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.2295/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>49212</start>
            <end>49228</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Reformulating Radiology Report Error Detection as Sentence-Level Vision-Language Learning
%A Min, Dabin
%A Jeon, Kyungmin
%A Lee, Dongheon
%A Park, Chang Min
%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 min-etal-2026-reformulating
%X Accurate alignment between chest X-ray (CXR) images and radiology reports is essential for clinical safety, yet automatic radiology report error detection from paired CXRs remains challenging. Prior approaches typically compare the entire report with the image, limiting the model’s ability to learn fine-grained vision-language correspondences at the level of individual findings. We reformulate report error detection as a sentence-level vision-language learning problem by decomposing reports into atomic finding statements, thereby reducing semantic entanglement and supervision noise. We propose SenRED (Sentence-level Report Error Detector), which uses sentence-conditioned patch aggregation to associate each statement with localized visual evidence. To supervise learning, we generate sentence-level synthetic errors reflecting clinically relevant categories—false positives, false negatives, and mislocalization—along with consistent image-sentence pairs. Experiments on MIMIC-CXR and Open-i show that SenRED consistently outperforms task-specific baselines and zero-shot instruction-tuned VLMs in both sentence-level and exam-level evaluations. Radiologist-confirmed evaluations further show that these gains are maintained on clinically reviewed subsets. Comparative experiments support the benefit of sentence-level reformulation, while ablations demonstrate that sentence-conditioned patch aggregation improves fine-grained consistency checking.
%R 10.18653/2026.findings-emnlp.2295
%U https://aclanthology.org/2026.findings-emnlp.2295/
%U https://doi.org/10.18653/2026.findings-emnlp.2295
%P 49212-49228

Markdown (Informal)

[Reformulating Radiology Report Error Detection as Sentence-Level Vision-Language Learning](https://aclanthology.org/2026.findings-emnlp.2295/) (Min et al., Findings 2026)

• Reformulating Radiology Report Error Detection as Sentence-Level Vision-Language Learning (Min et al., Findings 2026)

ACL

• Dabin Min, Kyungmin Jeon, Dongheon Lee, and Chang Min Park. 2026. Reformulating Radiology Report Error Detection as Sentence-Level Vision-Language Learning. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 49212–49228, Budapest, Hungary. Association for Computational Linguistics.