@inproceedings{lowphansirikul-ittichaiwong-2026-otter,
title = "Otter at {M}ed{E}x{A}ct2026: Diverse Encoder Ensemble for Medical Decision Span Detection",
author = "Lowphansirikul, Lalita and
Ittichaiwong, Piyalitt",
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-2.7/",
pages = "47--53",
ISBN = "979-8-89176-435-4",
abstract = "We build an ensemble of 10 transformer encoders for the MedExACT 2026 shared task on medical decision span detection. The ensemble is diversified along three training directions: encoder initialization (including domain-adaptive pre-training on clinical text), loss function, and data augmentation with LLM-generated synthetic notes and silver-labeled clinical documents. Greedy forward search selects the combination with the highest validation final score. A BERT-based boundary refiner is applied to the ensemble{'}s predicted spans to correct offset errors before submission."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lowphansirikul-ittichaiwong-2026-otter">
<titleInfo>
<title>Otter at MedExAct2026: Diverse Encoder Ensemble for Medical Decision Span Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lalita</namePart>
<namePart type="family">Lowphansirikul</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Piyalitt</namePart>
<namePart type="family">Ittichaiwong</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>We build an ensemble of 10 transformer encoders for the MedExACT 2026 shared task on medical decision span detection. The ensemble is diversified along three training directions: encoder initialization (including domain-adaptive pre-training on clinical text), loss function, and data augmentation with LLM-generated synthetic notes and silver-labeled clinical documents. Greedy forward search selects the combination with the highest validation final score. A BERT-based boundary refiner is applied to the ensemble’s predicted spans to correct offset errors before submission.</abstract>
<identifier type="citekey">lowphansirikul-ittichaiwong-2026-otter</identifier>
<location>
<url>https://aclanthology.org/2026.bionlp-2.7/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>47</start>
<end>53</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Otter at MedExAct2026: Diverse Encoder Ensemble for Medical Decision Span Detection
%A Lowphansirikul, Lalita
%A Ittichaiwong, Piyalitt
%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 lowphansirikul-ittichaiwong-2026-otter
%X We build an ensemble of 10 transformer encoders for the MedExACT 2026 shared task on medical decision span detection. The ensemble is diversified along three training directions: encoder initialization (including domain-adaptive pre-training on clinical text), loss function, and data augmentation with LLM-generated synthetic notes and silver-labeled clinical documents. Greedy forward search selects the combination with the highest validation final score. A BERT-based boundary refiner is applied to the ensemble’s predicted spans to correct offset errors before submission.
%U https://aclanthology.org/2026.bionlp-2.7/
%P 47-53
Markdown (Informal)
[Otter at MedExAct2026: Diverse Encoder Ensemble for Medical Decision Span Detection](https://aclanthology.org/2026.bionlp-2.7/) (Lowphansirikul & Ittichaiwong, BioNLP 2026)
ACL