<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-liu-2026-lotus">
    <titleInfo>
        <title>LOTUS: Language-Vision Optimal Transport for Unified Report Synthesis</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Xuhui</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jiachen</namePart>
        <namePart type="family">Liu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yue</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ruifeng</namePart>
            <namePart type="family">Xu</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ayu</namePart>
            <namePart type="family">Purwarianti</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tong</namePart>
            <namePart type="family">Rong</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jun</namePart>
            <namePart type="family">Suzuki</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Hengqin, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-444-6</identifier>
    </relatedItem>
    <abstract>Automated radiology report generation can reduce reporting burden, but existing encoder-decoder systems still rely on soft cross-attention that offers no geometric guarantee that generated tokens consume the correct visual evidence. We present LOTUS, an optimal-transport framework that places alignment supervision directly at the decoder evidence interface. LOTUS introduces Cross-Attention OT (CA-OT) to align decoder token states with visual patches, and extends it with Prototype OT (Proto-OT) for structure-level alignment between visual and textual codebooks. Experiments on IU X-Ray and ROCO, complemented by a scalability study on MIMIC-CXR-RRG, show that decoder-side OT improves macro-level clinical-label agreement while maintaining competitive language quality. In multi-seed IU X-Ray evaluation, LOTUS-Deep improves macro clinical F1 from 0.162 to 0.201 over the CE baseline, and on ROCO it yields micro-F1 gains on a thoracic-label proxy (0.395 to 0.600). We also provide a population-level risk-gap bound that justifies expected OT regularization as a principled cross-modal training signal.</abstract>
    <identifier type="citekey">li-liu-2026-lotus</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-aacl.86/</url>
    </location>
    <part>
        <date>2026-11</date>
        <extent unit="page">
            <start>1746</start>
            <end>1762</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T LOTUS: Language-Vision Optimal Transport for Unified Report Synthesis
%A Li, Xuhui
%A Liu, Jiachen
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-444-6
%F li-liu-2026-lotus
%X Automated radiology report generation can reduce reporting burden, but existing encoder-decoder systems still rely on soft cross-attention that offers no geometric guarantee that generated tokens consume the correct visual evidence. We present LOTUS, an optimal-transport framework that places alignment supervision directly at the decoder evidence interface. LOTUS introduces Cross-Attention OT (CA-OT) to align decoder token states with visual patches, and extends it with Prototype OT (Proto-OT) for structure-level alignment between visual and textual codebooks. Experiments on IU X-Ray and ROCO, complemented by a scalability study on MIMIC-CXR-RRG, show that decoder-side OT improves macro-level clinical-label agreement while maintaining competitive language quality. In multi-seed IU X-Ray evaluation, LOTUS-Deep improves macro clinical F1 from 0.162 to 0.201 over the CE baseline, and on ROCO it yields micro-F1 gains on a thoracic-label proxy (0.395 to 0.600). We also provide a population-level risk-gap bound that justifies expected OT regularization as a principled cross-modal training signal.
%U https://aclanthology.org/2026.findings-aacl.86/
%P 1746-1762

Markdown (Informal)

[LOTUS: Language-Vision Optimal Transport for Unified Report Synthesis](https://aclanthology.org/2026.findings-aacl.86/) (Li & Liu, Findings 2026)

• LOTUS: Language-Vision Optimal Transport for Unified Report Synthesis (Li & Liu, Findings 2026)

ACL

• Xuhui Li and Jiachen Liu. 2026. LOTUS: Language-Vision Optimal Transport for Unified Report Synthesis. In Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026, pages 1746–1762, Hengqin, China. Association for Computational Linguistics.