<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-yang-2026-njust">
    <titleInfo>
        <title>NJUST-KMG at MedGenVidQA 2026: Cascade Multi-modal Alignment with Gaussian Soft Priors for Medical Visual Answer Localization</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Jinglong</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yang</namePart>
        <namePart type="family">Yang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the BioNLP 2026 (Shared Tasks)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Deepak</namePart>
            <namePart type="family">Gupta</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Dina</namePart>
            <namePart type="family">Demner-Fushman</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">San Diego, California, USA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-435-4</identifier>
    </relatedItem>
    <abstract>This paper describes the system developed for the Medical Visual Answer Localization (MVAL) task at MedGenVidQA 2026. Accurately locating surgical or instructional steps in medical videos is inherently challenging due to audio-visual asynchrony and the visual homogeneity of surgical scenes. We propose a Cascade Multi-modal Alignment Framework that integrates Large Language Models (LLMs) to bridge the semantic-temporal gap. Our pipeline utilizes WhisperX for word-level speech transcription to ensure precise textual anchoring. We then employ Gemini3 as a high-level semantic ranker to generate multi-scale textual priors. Crucially, we transform these discrete semantic scores into a continuous 1D Gaussian Soft Prior, which is injected as an attention bias into our cross-modal fusion network. This mechanism preserves global temporal context while guiding the model to focus on query-relevant frames. Our system achieves highly competitive performance on the validation leaderboard, particularly under strict evaluation metrics, reaching an IoU@0.7 of 67.5%.</abstract>
    <identifier type="citekey">li-yang-2026-njust</identifier>
    <location>
        <url>https://aclanthology.org/2026.bionlp-2.30/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>229</start>
            <end>232</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T NJUST-KMG at MedGenVidQA 2026: Cascade Multi-modal Alignment with Gaussian Soft Priors for Medical Visual Answer Localization
%A Li, Jinglong
%A Yang, Yang
%Y Gupta, Deepak
%Y Demner-Fushman, Dina
%S Proceedings of the BioNLP 2026 (Shared Tasks)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, USA
%@ 979-8-89176-435-4
%F li-yang-2026-njust
%X This paper describes the system developed for the Medical Visual Answer Localization (MVAL) task at MedGenVidQA 2026. Accurately locating surgical or instructional steps in medical videos is inherently challenging due to audio-visual asynchrony and the visual homogeneity of surgical scenes. We propose a Cascade Multi-modal Alignment Framework that integrates Large Language Models (LLMs) to bridge the semantic-temporal gap. Our pipeline utilizes WhisperX for word-level speech transcription to ensure precise textual anchoring. We then employ Gemini3 as a high-level semantic ranker to generate multi-scale textual priors. Crucially, we transform these discrete semantic scores into a continuous 1D Gaussian Soft Prior, which is injected as an attention bias into our cross-modal fusion network. This mechanism preserves global temporal context while guiding the model to focus on query-relevant frames. Our system achieves highly competitive performance on the validation leaderboard, particularly under strict evaluation metrics, reaching an IoU@0.7 of 67.5%.
%U https://aclanthology.org/2026.bionlp-2.30/
%P 229-232

Markdown (Informal)

[NJUST-KMG at MedGenVidQA 2026: Cascade Multi-modal Alignment with Gaussian Soft Priors for Medical Visual Answer Localization](https://aclanthology.org/2026.bionlp-2.30/) (Li & Yang, BioNLP 2026)

• NJUST-KMG at MedGenVidQA 2026: Cascade Multi-modal Alignment with Gaussian Soft Priors for Medical Visual Answer Localization (Li & Yang, BioNLP 2026)

ACL

• Jinglong Li and Yang Yang. 2026. NJUST-KMG at MedGenVidQA 2026: Cascade Multi-modal Alignment with Gaussian Soft Priors for Medical Visual Answer Localization. In Proceedings of the BioNLP 2026 (Shared Tasks), pages 229–232, San Diego, California, USA. Association for Computational Linguistics.