@inproceedings{hwang-etal-2026-semantica,
title = "{S}em{A}n{TICA} Lab at {M}edi{QA}-{SYNUR} 2026: Route, Extract and Verify {--} An {LLM}-gated Ensemble for Parsing Nurse Dictations",
author = "Hwang, Sy and
Pitcher, Katherine S. and
Kim, Sue Hyon and
Lee, Yoonjae and
Donelly, Hayoung K. and
Bandhey, Harsh and
King, Andrew J. and
O{'}Connor, Karen and
Urbanowicz, Ryan J. and
Mowery, Danielle L.",
editor = "Ben Abacha, Asma and
Bethard, Steven and
Bitterman, Danielle and
Naumann, Tristan and
Roberts, Kirk",
booktitle = "Proceedings of the 8th Workshop on Clinical Natural Language Processing (Clinical {NLP}) @ {LREC} 2026",
month = may,
year = "2026",
address = "Palma, Mallorca (Spain)",
publisher = "ELRA Language Resources Association (ELRA)",
url = "https://aclanthology.org/2026.clinicalnlp-1.4/",
doi = "10.63317/3cfpmqqpmdq5",
pages = "27--37",
abstract = "We describe the Semantic Analysis of Text to Inform Clinical Action (SemAnTICA) Lab{'}s system for the MediQA-SYNUR 2026 shared task on extracting structured clinical observations from nurse dictation transcripts. The task requires mapping observations from disfluent conversational text to a large, fixed ontology and producing strictly normalized outputs, where small amounts of concept over-selection severely degrade micro-F1 score. Our approach evolved from a full-schema in-context baseline to a pipeline that explicitly separates concept selection from value extraction. We first preprocess transcripts, then generate transcript-specific concept candidates using hybrid sparse{--}dense retrieval. The candidates are then pruned with an evidence-based filter. For extraction, we adopt a system-level mixture-of-experts design with an online LLM router that selects a subset of domain-specialized experts per transcript. Each expert operates over a constrained schema partition to reduce spurious predictions. We enhance robustness with agreement-gated ensembling and targeted adjudication for ambiguous cases. Finally, we intersect complementary high-recall and high-precision runs to produce the best submission. Our system ranked first on the official test leaderboard with F1 = 0.814, P = 0.826, R = 0.801."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hwang-etal-2026-semantica">
<titleInfo>
<title>SemAnTICA Lab at MediQA-SYNUR 2026: Route, Extract and Verify – An LLM-gated Ensemble for Parsing Nurse Dictations</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sy</namePart>
<namePart type="family">Hwang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Katherine</namePart>
<namePart type="given">S</namePart>
<namePart type="family">Pitcher</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sue</namePart>
<namePart type="given">Hyon</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yoonjae</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hayoung</namePart>
<namePart type="given">K</namePart>
<namePart type="family">Donelly</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Harsh</namePart>
<namePart type="family">Bandhey</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrew</namePart>
<namePart type="given">J</namePart>
<namePart type="family">King</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Karen</namePart>
<namePart type="family">O’Connor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ryan</namePart>
<namePart type="given">J</namePart>
<namePart type="family">Urbanowicz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Danielle</namePart>
<namePart type="given">L</namePart>
<namePart type="family">Mowery</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 8th Workshop on Clinical Natural Language Processing (Clinical NLP) @ LREC 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Asma</namePart>
<namePart type="family">Ben Abacha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Steven</namePart>
<namePart type="family">Bethard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Danielle</namePart>
<namePart type="family">Bitterman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tristan</namePart>
<namePart type="family">Naumann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kirk</namePart>
<namePart type="family">Roberts</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Palma, Mallorca (Spain)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We describe the Semantic Analysis of Text to Inform Clinical Action (SemAnTICA) Lab’s system for the MediQA-SYNUR 2026 shared task on extracting structured clinical observations from nurse dictation transcripts. The task requires mapping observations from disfluent conversational text to a large, fixed ontology and producing strictly normalized outputs, where small amounts of concept over-selection severely degrade micro-F1 score. Our approach evolved from a full-schema in-context baseline to a pipeline that explicitly separates concept selection from value extraction. We first preprocess transcripts, then generate transcript-specific concept candidates using hybrid sparse–dense retrieval. The candidates are then pruned with an evidence-based filter. For extraction, we adopt a system-level mixture-of-experts design with an online LLM router that selects a subset of domain-specialized experts per transcript. Each expert operates over a constrained schema partition to reduce spurious predictions. We enhance robustness with agreement-gated ensembling and targeted adjudication for ambiguous cases. Finally, we intersect complementary high-recall and high-precision runs to produce the best submission. Our system ranked first on the official test leaderboard with F1 = 0.814, P = 0.826, R = 0.801.</abstract>
<identifier type="citekey">hwang-etal-2026-semantica</identifier>
<identifier type="doi">10.63317/3cfpmqqpmdq5</identifier>
<location>
<url>https://aclanthology.org/2026.clinicalnlp-1.4/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>27</start>
<end>37</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SemAnTICA Lab at MediQA-SYNUR 2026: Route, Extract and Verify – An LLM-gated Ensemble for Parsing Nurse Dictations
%A Hwang, Sy
%A Pitcher, Katherine S.
%A Kim, Sue Hyon
%A Lee, Yoonjae
%A Donelly, Hayoung K.
%A Bandhey, Harsh
%A King, Andrew J.
%A O’Connor, Karen
%A Urbanowicz, Ryan J.
%A Mowery, Danielle L.
%Y Ben Abacha, Asma
%Y Bethard, Steven
%Y Bitterman, Danielle
%Y Naumann, Tristan
%Y Roberts, Kirk
%S Proceedings of the 8th Workshop on Clinical Natural Language Processing (Clinical NLP) @ LREC 2026
%D 2026
%8 May
%I ELRA Language Resources Association (ELRA)
%C Palma, Mallorca (Spain)
%F hwang-etal-2026-semantica
%X We describe the Semantic Analysis of Text to Inform Clinical Action (SemAnTICA) Lab’s system for the MediQA-SYNUR 2026 shared task on extracting structured clinical observations from nurse dictation transcripts. The task requires mapping observations from disfluent conversational text to a large, fixed ontology and producing strictly normalized outputs, where small amounts of concept over-selection severely degrade micro-F1 score. Our approach evolved from a full-schema in-context baseline to a pipeline that explicitly separates concept selection from value extraction. We first preprocess transcripts, then generate transcript-specific concept candidates using hybrid sparse–dense retrieval. The candidates are then pruned with an evidence-based filter. For extraction, we adopt a system-level mixture-of-experts design with an online LLM router that selects a subset of domain-specialized experts per transcript. Each expert operates over a constrained schema partition to reduce spurious predictions. We enhance robustness with agreement-gated ensembling and targeted adjudication for ambiguous cases. Finally, we intersect complementary high-recall and high-precision runs to produce the best submission. Our system ranked first on the official test leaderboard with F1 = 0.814, P = 0.826, R = 0.801.
%R 10.63317/3cfpmqqpmdq5
%U https://aclanthology.org/2026.clinicalnlp-1.4/
%U https://doi.org/10.63317/3cfpmqqpmdq5
%P 27-37
Markdown (Informal)
[SemAnTICA Lab at MediQA-SYNUR 2026: Route, Extract and Verify – An LLM-gated Ensemble for Parsing Nurse Dictations](https://aclanthology.org/2026.clinicalnlp-1.4/) (Hwang et al., ClinicalNLP 2026)
ACL
- Sy Hwang, Katherine S. Pitcher, Sue Hyon Kim, Yoonjae Lee, Hayoung K. Donelly, Harsh Bandhey, Andrew J. King, Karen O’Connor, Ryan J. Urbanowicz, and Danielle L. Mowery. 2026. SemAnTICA Lab at MediQA-SYNUR 2026: Route, Extract and Verify – An LLM-gated Ensemble for Parsing Nurse Dictations. In Proceedings of the 8th Workshop on Clinical Natural Language Processing (Clinical NLP) @ LREC 2026, pages 27–37, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).