@inproceedings{li-etal-2026-dlnlp,
title = "{DLNLP} at {C}linical{S}kill{QA}: {E}vidence{F}low for Structured Zero-Shot Clinical Keyframe Ordering",
author = "Li, Kexin and
Wang, Zhekun and
Wang, Yiran and
Zhao, Di",
editor = "Gupta, Deepak and
Demner-Fushman, Dina",
booktitle = "Proceedings of the {B}io{NLP} 2026 (Shared Tasks)",
month = jul,
year = "2026",
address = "San Diego, California, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.bionlp-2.5/",
pages = "33--37",
ISBN = "979-8-89176-435-4",
abstract = "The ClinSkill QA shared task requires models to recover the temporal order of scrambled clinical keyframes and generate explanations. We propose EvidenceFlow, a structured zero-shot framework based on Qwen2.5-VL that decomposes the task into global overview, local evidence modeling, and ordering decision, with two variants: model-led EvidenceFlow-M and rule-guided EvidenceFlow-R. On the official test set, EvidenceFlow-R achieves better ordering performance, while EvidenceFlow-M produces better explanation quality, revealing a trade-off between ordering stability and rationale generation. EvidenceFlow provides an interpretable zero-shot baseline for clinical keyframe ordering."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2026-dlnlp">
<titleInfo>
<title>DLNLP at ClinicalSkillQA: EvidenceFlow for Structured Zero-Shot Clinical Keyframe Ordering</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kexin</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhekun</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yiran</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Di</namePart>
<namePart type="family">Zhao</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>The ClinSkill QA shared task requires models to recover the temporal order of scrambled clinical keyframes and generate explanations. We propose EvidenceFlow, a structured zero-shot framework based on Qwen2.5-VL that decomposes the task into global overview, local evidence modeling, and ordering decision, with two variants: model-led EvidenceFlow-M and rule-guided EvidenceFlow-R. On the official test set, EvidenceFlow-R achieves better ordering performance, while EvidenceFlow-M produces better explanation quality, revealing a trade-off between ordering stability and rationale generation. EvidenceFlow provides an interpretable zero-shot baseline for clinical keyframe ordering.</abstract>
<identifier type="citekey">li-etal-2026-dlnlp</identifier>
<location>
<url>https://aclanthology.org/2026.bionlp-2.5/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>33</start>
<end>37</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T DLNLP at ClinicalSkillQA: EvidenceFlow for Structured Zero-Shot Clinical Keyframe Ordering
%A Li, Kexin
%A Wang, Zhekun
%A Wang, Yiran
%A Zhao, Di
%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-etal-2026-dlnlp
%X The ClinSkill QA shared task requires models to recover the temporal order of scrambled clinical keyframes and generate explanations. We propose EvidenceFlow, a structured zero-shot framework based on Qwen2.5-VL that decomposes the task into global overview, local evidence modeling, and ordering decision, with two variants: model-led EvidenceFlow-M and rule-guided EvidenceFlow-R. On the official test set, EvidenceFlow-R achieves better ordering performance, while EvidenceFlow-M produces better explanation quality, revealing a trade-off between ordering stability and rationale generation. EvidenceFlow provides an interpretable zero-shot baseline for clinical keyframe ordering.
%U https://aclanthology.org/2026.bionlp-2.5/
%P 33-37
Markdown (Informal)
[DLNLP at ClinicalSkillQA: EvidenceFlow for Structured Zero-Shot Clinical Keyframe Ordering](https://aclanthology.org/2026.bionlp-2.5/) (Li et al., BioNLP 2026)
ACL