@inproceedings{shao-etal-2026-disagreement,
title = "Disagreement-Driven Joint Refinement of Retrieval and Decision Rules for Imbalanced Counseling Risk Classification",
author = "Shao, Zhihao and
Sekizaki, Ryo and
Yi, Shengzhou and
Yamasaki, Toshihiko",
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-1.32/",
doi = "10.63317/2f8zttankw7p",
pages = "281--289",
abstract = "With the rapid growth of online counseling services, timely and reliable risk classification of counseling records is essential for supporting early screening and prioritizing limited intervention resources. High-risk samples refer to high-acuity suicide risk and require expedited human review. However, this task is challenging due to severe class imbalance (93{\%} low-risk and 7{\%} high-risk samples) and complex decision boundaries. Large language models (LLMs) exhibit unstable predictions and systematic errors in such imbalanced clinical-text settings. To address this issue, we propose Disagreement-Driven Joint Refinement (DDJR), an iterative, parameter-free refinement framework. It uses prediction disagreement between two inference settings, zero-shot and retrieval-augmented in-context learning, as the primary signal for identifying high-value instances. These disagreement-identified instances are transformed into adaptive refinement signals and used to jointly update both the exemplar pool and an executable rule set, thereby sharpening decision boundaries and improving prediction stability. Experiments on 6,481 real-world counseling records demonstrate that the proposed DDJR outperforms existing methods, achieving an accuracy of 0.915 and a Matthews Correlation Coefficient (MCC) of 0.583. These results demonstrate that DDJR achieves more stable and reliable predictions for high-stakes counseling risk classification in real-world settings."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shao-etal-2026-disagreement">
<titleInfo>
<title>Disagreement-Driven Joint Refinement of Retrieval and Decision Rules for Imbalanced Counseling Risk Classification</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhihao</namePart>
<namePart type="family">Shao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ryo</namePart>
<namePart type="family">Sekizaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shengzhou</namePart>
<namePart type="family">Yi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Toshihiko</namePart>
<namePart type="family">Yamasaki</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>With the rapid growth of online counseling services, timely and reliable risk classification of counseling records is essential for supporting early screening and prioritizing limited intervention resources. High-risk samples refer to high-acuity suicide risk and require expedited human review. However, this task is challenging due to severe class imbalance (93% low-risk and 7% high-risk samples) and complex decision boundaries. Large language models (LLMs) exhibit unstable predictions and systematic errors in such imbalanced clinical-text settings. To address this issue, we propose Disagreement-Driven Joint Refinement (DDJR), an iterative, parameter-free refinement framework. It uses prediction disagreement between two inference settings, zero-shot and retrieval-augmented in-context learning, as the primary signal for identifying high-value instances. These disagreement-identified instances are transformed into adaptive refinement signals and used to jointly update both the exemplar pool and an executable rule set, thereby sharpening decision boundaries and improving prediction stability. Experiments on 6,481 real-world counseling records demonstrate that the proposed DDJR outperforms existing methods, achieving an accuracy of 0.915 and a Matthews Correlation Coefficient (MCC) of 0.583. These results demonstrate that DDJR achieves more stable and reliable predictions for high-stakes counseling risk classification in real-world settings.</abstract>
<identifier type="citekey">shao-etal-2026-disagreement</identifier>
<identifier type="doi">10.63317/2f8zttankw7p</identifier>
<location>
<url>https://aclanthology.org/2026.clinicalnlp-1.32/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>281</start>
<end>289</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Disagreement-Driven Joint Refinement of Retrieval and Decision Rules for Imbalanced Counseling Risk Classification
%A Shao, Zhihao
%A Sekizaki, Ryo
%A Yi, Shengzhou
%A Yamasaki, Toshihiko
%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 shao-etal-2026-disagreement
%X With the rapid growth of online counseling services, timely and reliable risk classification of counseling records is essential for supporting early screening and prioritizing limited intervention resources. High-risk samples refer to high-acuity suicide risk and require expedited human review. However, this task is challenging due to severe class imbalance (93% low-risk and 7% high-risk samples) and complex decision boundaries. Large language models (LLMs) exhibit unstable predictions and systematic errors in such imbalanced clinical-text settings. To address this issue, we propose Disagreement-Driven Joint Refinement (DDJR), an iterative, parameter-free refinement framework. It uses prediction disagreement between two inference settings, zero-shot and retrieval-augmented in-context learning, as the primary signal for identifying high-value instances. These disagreement-identified instances are transformed into adaptive refinement signals and used to jointly update both the exemplar pool and an executable rule set, thereby sharpening decision boundaries and improving prediction stability. Experiments on 6,481 real-world counseling records demonstrate that the proposed DDJR outperforms existing methods, achieving an accuracy of 0.915 and a Matthews Correlation Coefficient (MCC) of 0.583. These results demonstrate that DDJR achieves more stable and reliable predictions for high-stakes counseling risk classification in real-world settings.
%R 10.63317/2f8zttankw7p
%U https://aclanthology.org/2026.clinicalnlp-1.32/
%U https://doi.org/10.63317/2f8zttankw7p
%P 281-289
Markdown (Informal)
[Disagreement-Driven Joint Refinement of Retrieval and Decision Rules for Imbalanced Counseling Risk Classification](https://aclanthology.org/2026.clinicalnlp-1.32/) (Shao et al., ClinicalNLP 2026)
ACL