<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="nahian-kavuluru-2025-radqa">
    <titleInfo>
        <title>RadQA-DPO: A Radiology Question Answering System with Encoder-Decoder Models Enhanced by Direct Preference Optimization</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Md</namePart>
        <namePart type="given">Sultan</namePart>
        <namePart type="given">Al</namePart>
        <namePart type="family">Nahian</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ramakanth</namePart>
        <namePart type="family">Kavuluru</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-08</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 24th Workshop on Biomedical Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Dina</namePart>
            <namePart type="family">Demner-Fushman</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sophia</namePart>
            <namePart type="family">Ananiadou</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Makoto</namePart>
            <namePart type="family">Miwa</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Junichi</namePart>
            <namePart type="family">Tsujii</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Viena, Austria</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-275-6</identifier>
    </relatedItem>
    <abstract>Extractive question answering over clinical text is a crucial need to help deal with the deluge of clinical text generated in hospitals. While encoder models (e.g., BERT) have been popular for this reading comprehension–style question answering task, recently encoder-decoder models (e.g., T5) are on the rise. There is also the emergence of preference optimization techniques to align decoder-only LLMs with human preferences. In this paper, we combine encoder-decoder models with the direct preference optimization (DPO) method for the RadQA radiology question answering task. Our approach achieves a 12–15 F1 point improvement over previous state-of-the-art models. To the best of our knowledge, this effort is the first to show that DPO method also works for reading comprehension via novel heuristics to generate preference data without human inputs.</abstract>
    <identifier type="citekey">nahian-kavuluru-2025-radqa</identifier>
    <identifier type="doi">10.18653/v1/2025.bionlp-1.10</identifier>
    <location>
        <url>https://aclanthology.org/2025.bionlp-1.10/</url>
    </location>
    <part>
        <date>2025-08</date>
        <extent unit="page">
            <start>101</start>
            <end>113</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T RadQA-DPO: A Radiology Question Answering System with Encoder-Decoder Models Enhanced by Direct Preference Optimization
%A Nahian, Md Sultan Al
%A Kavuluru, Ramakanth
%Y Demner-Fushman, Dina
%Y Ananiadou, Sophia
%Y Miwa, Makoto
%Y Tsujii, Junichi
%S Proceedings of the 24th Workshop on Biomedical Language Processing
%D 2025
%8 August
%I Association for Computational Linguistics
%C Viena, Austria
%@ 979-8-89176-275-6
%F nahian-kavuluru-2025-radqa
%X Extractive question answering over clinical text is a crucial need to help deal with the deluge of clinical text generated in hospitals. While encoder models (e.g., BERT) have been popular for this reading comprehension–style question answering task, recently encoder-decoder models (e.g., T5) are on the rise. There is also the emergence of preference optimization techniques to align decoder-only LLMs with human preferences. In this paper, we combine encoder-decoder models with the direct preference optimization (DPO) method for the RadQA radiology question answering task. Our approach achieves a 12–15 F1 point improvement over previous state-of-the-art models. To the best of our knowledge, this effort is the first to show that DPO method also works for reading comprehension via novel heuristics to generate preference data without human inputs.
%R 10.18653/v1/2025.bionlp-1.10
%U https://aclanthology.org/2025.bionlp-1.10/
%U https://doi.org/10.18653/v1/2025.bionlp-1.10
%P 101-113

Markdown (Informal)

[RadQA-DPO: A Radiology Question Answering System with Encoder-Decoder Models Enhanced by Direct Preference Optimization](https://aclanthology.org/2025.bionlp-1.10/) (Nahian & Kavuluru, BioNLP 2025)

• RadQA-DPO: A Radiology Question Answering System with Encoder-Decoder Models Enhanced by Direct Preference Optimization (Nahian & Kavuluru, BioNLP 2025)

ACL

• Md Sultan Al Nahian and Ramakanth Kavuluru. 2025. RadQA-DPO: A Radiology Question Answering System with Encoder-Decoder Models Enhanced by Direct Preference Optimization. In Proceedings of the 24th Workshop on Biomedical Language Processing, pages 101–113, Viena, Austria. Association for Computational Linguistics.