@inproceedings{li-etal-2026-pediatric,
title = "Pediatric Sepsis Cohort Detection Using In-Context Pointwise {V}-Usable Information",
author = "Li, Yingya and
Geva, Alon and
Bethard, Steven and
Miller, Timothy A. and
Madden, Kate and
Eisenberg, Matthew A. and
Kelly, Daniel P. and
Savova, Guergana",
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-1.37/",
doi = "10.63317/38bj4pwcnt3q",
pages = "336--349",
abstract = "Pediatric sepsis diagnosis remains a major clinical challenge due to non-specific symptoms and a lack of reliable diagnostic criteria. Large language models (LLMs) provide a scalable solution for processing and understanding unstructured text in medical records. However, identifying the most suitable model is non-trivial given the rapid growth of available LLMs. In this work, we proposed using in-context pointwise V-usable information (pvi) to estimate task difficulty and guide model selection for pediatric sepsis cohort detection. We applied in-context pvi to estimate task difficulty and inform model selection across 12 state-of-the-art open LLMs on the task, using electronic medical record data from 507 patient encounters at a U.S. children{'}s hospital. We compared the performance of the best-fitting LLM to feature-rich baseline models and a fine-tuned transformer. Our results show that the pvi-selected LLM outperforms the baselines, although the feature-rich bag-of-words model with a support vector machine also achieves competitive performance. We believe our approach demonstrates a promising application of current LLM techniques to high-stakes clinical tasks."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2026-pediatric">
<titleInfo>
<title>Pediatric Sepsis Cohort Detection Using In-Context Pointwise V-Usable Information</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yingya</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alon</namePart>
<namePart type="family">Geva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Steven</namePart>
<namePart type="family">Bethard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Timothy</namePart>
<namePart type="given">A</namePart>
<namePart type="family">Miller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kate</namePart>
<namePart type="family">Madden</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matthew</namePart>
<namePart type="given">A</namePart>
<namePart type="family">Eisenberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Kelly</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Guergana</namePart>
<namePart type="family">Savova</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>Pediatric sepsis diagnosis remains a major clinical challenge due to non-specific symptoms and a lack of reliable diagnostic criteria. Large language models (LLMs) provide a scalable solution for processing and understanding unstructured text in medical records. However, identifying the most suitable model is non-trivial given the rapid growth of available LLMs. In this work, we proposed using in-context pointwise V-usable information (pvi) to estimate task difficulty and guide model selection for pediatric sepsis cohort detection. We applied in-context pvi to estimate task difficulty and inform model selection across 12 state-of-the-art open LLMs on the task, using electronic medical record data from 507 patient encounters at a U.S. children’s hospital. We compared the performance of the best-fitting LLM to feature-rich baseline models and a fine-tuned transformer. Our results show that the pvi-selected LLM outperforms the baselines, although the feature-rich bag-of-words model with a support vector machine also achieves competitive performance. We believe our approach demonstrates a promising application of current LLM techniques to high-stakes clinical tasks.</abstract>
<identifier type="citekey">li-etal-2026-pediatric</identifier>
<identifier type="doi">10.63317/38bj4pwcnt3q</identifier>
<location>
<url>https://aclanthology.org/2026.clinicalnlp-1.37/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>336</start>
<end>349</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Pediatric Sepsis Cohort Detection Using In-Context Pointwise V-Usable Information
%A Li, Yingya
%A Geva, Alon
%A Bethard, Steven
%A Miller, Timothy A.
%A Madden, Kate
%A Eisenberg, Matthew A.
%A Kelly, Daniel P.
%A Savova, Guergana
%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 li-etal-2026-pediatric
%X Pediatric sepsis diagnosis remains a major clinical challenge due to non-specific symptoms and a lack of reliable diagnostic criteria. Large language models (LLMs) provide a scalable solution for processing and understanding unstructured text in medical records. However, identifying the most suitable model is non-trivial given the rapid growth of available LLMs. In this work, we proposed using in-context pointwise V-usable information (pvi) to estimate task difficulty and guide model selection for pediatric sepsis cohort detection. We applied in-context pvi to estimate task difficulty and inform model selection across 12 state-of-the-art open LLMs on the task, using electronic medical record data from 507 patient encounters at a U.S. children’s hospital. We compared the performance of the best-fitting LLM to feature-rich baseline models and a fine-tuned transformer. Our results show that the pvi-selected LLM outperforms the baselines, although the feature-rich bag-of-words model with a support vector machine also achieves competitive performance. We believe our approach demonstrates a promising application of current LLM techniques to high-stakes clinical tasks.
%R 10.63317/38bj4pwcnt3q
%U https://aclanthology.org/2026.clinicalnlp-1.37/
%U https://doi.org/10.63317/38bj4pwcnt3q
%P 336-349
Markdown (Informal)
[Pediatric Sepsis Cohort Detection Using In-Context Pointwise V-Usable Information](https://aclanthology.org/2026.clinicalnlp-1.37/) (Li et al., ClinicalNLP 2026)
ACL
- Yingya Li, Alon Geva, Steven Bethard, Timothy A. Miller, Kate Madden, Matthew A. Eisenberg, Daniel P. Kelly, and Guergana Savova. 2026. Pediatric Sepsis Cohort Detection Using In-Context Pointwise V-Usable Information. In Proceedings of the 8th Workshop on Clinical Natural Language Processing (Clinical NLP) @ LREC 2026, pages 336–349, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).