<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2025-docmmir">
    <titleInfo>
        <title>DocMMIR: A Framework for Document Multi-modal Information Retrieval</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Zirui</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Siwei</namePart>
        <namePart type="family">Wu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yizhi</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xingyu</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yi</namePart>
        <namePart type="family">Zhou</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Chenghua</namePart>
        <namePart type="family">Lin</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2025</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Christos</namePart>
            <namePart type="family">Christodoulopoulos</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tanmoy</namePart>
            <namePart type="family">Chakraborty</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Carolyn</namePart>
            <namePart type="family">Rose</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Violet</namePart>
            <namePart type="family">Peng</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Suzhou, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-335-7</identifier>
    </relatedItem>
    <abstract>The rapid advancement of unsupervised representation learning and large-scale pre-trained vision-language models has significantly improved cross-modal retrieval tasks. However, existing multi-modal information retrieval (MMIR) studies lack a comprehensive exploration of document-level retrieval and suffer from the absence of cross-domain datasets at this granularity. To address this limitation, we introduce DocMMIR, a novel multi-modal document retrieval framework designed explicitly to unify diverse document formats and domains—including Wikipedia articles, scientific papers (arXiv), and presentation slides—within a comprehensive retrieval scenario. We construct a large-scale cross-domain multimodal dataset, comprising 450K training, 19.2K validation, and 19.2K test documents, serving as both a benchmark to reveal the shortcomings of existing MMIR models and a training set for further improvement. The dataset systematically integrates textual and visual information. Our comprehensive experimental analysis reveals substantial limitations in current state-of-the-art MLLMs (CLIP, BLIP2, SigLIP-2, ALIGN) when applied to our tasks, with only CLIP (ViT-L/14) demonstrating reasonable zero-shot performance. Through systematic investigation of cross-modal fusion strategies and loss function selection on the CLIP (ViT-L/14) model, we develop an optimised approach that achieves a +31% improvement in MRR@10 metrics from zero-shot baseline to fine-tuned model. Our findings offer crucial insights and practical guidance for future development in unified multimodal document retrieval tasks.</abstract>
    <identifier type="citekey">li-etal-2025-docmmir</identifier>
    <location>
        <url>https://aclanthology.org/2025.findings-emnlp.705/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>13117</start>
            <end>13130</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T DocMMIR: A Framework for Document Multi-modal Information Retrieval
%A Li, Zirui
%A Wu, Siwei
%A Li, Yizhi
%A Wang, Xingyu
%A Zhou, Yi
%A Lin, Chenghua
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Findings of the Association for Computational Linguistics: EMNLP 2025
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-335-7
%F li-etal-2025-docmmir
%X The rapid advancement of unsupervised representation learning and large-scale pre-trained vision-language models has significantly improved cross-modal retrieval tasks. However, existing multi-modal information retrieval (MMIR) studies lack a comprehensive exploration of document-level retrieval and suffer from the absence of cross-domain datasets at this granularity. To address this limitation, we introduce DocMMIR, a novel multi-modal document retrieval framework designed explicitly to unify diverse document formats and domains—including Wikipedia articles, scientific papers (arXiv), and presentation slides—within a comprehensive retrieval scenario. We construct a large-scale cross-domain multimodal dataset, comprising 450K training, 19.2K validation, and 19.2K test documents, serving as both a benchmark to reveal the shortcomings of existing MMIR models and a training set for further improvement. The dataset systematically integrates textual and visual information. Our comprehensive experimental analysis reveals substantial limitations in current state-of-the-art MLLMs (CLIP, BLIP2, SigLIP-2, ALIGN) when applied to our tasks, with only CLIP (ViT-L/14) demonstrating reasonable zero-shot performance. Through systematic investigation of cross-modal fusion strategies and loss function selection on the CLIP (ViT-L/14) model, we develop an optimised approach that achieves a +31% improvement in MRR@10 metrics from zero-shot baseline to fine-tuned model. Our findings offer crucial insights and practical guidance for future development in unified multimodal document retrieval tasks.
%U https://aclanthology.org/2025.findings-emnlp.705/
%P 13117-13130

Markdown (Informal)

[DocMMIR: A Framework for Document Multi-modal Information Retrieval](https://aclanthology.org/2025.findings-emnlp.705/) (Li et al., Findings 2025)

• DocMMIR: A Framework for Document Multi-modal Information Retrieval (Li et al., Findings 2025)

ACL

• Zirui Li, Siwei Wu, Yizhi Li, Xingyu Wang, Yi Zhou, and Chenghua Lin. 2025. DocMMIR: A Framework for Document Multi-modal Information Retrieval. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 13117–13130, Suzhou, China. Association for Computational Linguistics.