@inproceedings{chen-etal-2026-identifying,
title = "Identifying Student Real Weaknesses: Learnable Error-Aware Distillation for Medical Question Answering",
author = "Chen, Yonghe and
Meng, Xiaoyang and
Zhou, Xiaoding and
Zhang, Wentian and
Zhao, Ruimeng and
Lian, Ying and
Chen, Guoqiang and
Zhang, Weiyu and
Lu, Wenpeng",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.569/",
pages = "11982--12004",
ISBN = "979-8-89176-462-0",
abstract = "Large language models (LLMs) show strong performance in medical question answering, but their computational costs limit deployment under resource constraints. Therefore, an increasing number of studies focus on distilling knowledge from LLMs into smaller models to improve performance in medical question answering. However, existing methods often fail to identify truly useful knowledge to distill, making it difficult to focus distillation on the student{'}s real weaknesses. As a result, much of the distillation budget is spent on already acquired knowledge rather than on weaknesses that still limit performance. To address this problem, we propose LEAD, a Learnable Error-Aware Distillation framework that precisely improves the weaknesses of student models. First, LEAD identifies questions answered incorrectly by the student model to determine which weaknesses require teacher guidance. Then, LEAD generates teacher rationales and similar samples to construct candidate supervision for these errors. Through supervision refinement, reliable and learnable training samples are retained from the candidate supervision. Finally, LEAD trains the student on a balanced sample mixture to correct learnable errors while preserving previously acquired knowledge. Experiments on four medical question answering benchmarks show that LEAD outperforms existing distillation methods."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chen-etal-2026-identifying">
<titleInfo>
<title>Identifying Student Real Weaknesses: Learnable Error-Aware Distillation for Medical Question Answering</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yonghe</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaoyang</namePart>
<namePart type="family">Meng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaoding</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wentian</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruimeng</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ying</namePart>
<namePart type="family">Lian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Guoqiang</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weiyu</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenpeng</namePart>
<namePart type="family">Lu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Large language models (LLMs) show strong performance in medical question answering, but their computational costs limit deployment under resource constraints. Therefore, an increasing number of studies focus on distilling knowledge from LLMs into smaller models to improve performance in medical question answering. However, existing methods often fail to identify truly useful knowledge to distill, making it difficult to focus distillation on the student’s real weaknesses. As a result, much of the distillation budget is spent on already acquired knowledge rather than on weaknesses that still limit performance. To address this problem, we propose LEAD, a Learnable Error-Aware Distillation framework that precisely improves the weaknesses of student models. First, LEAD identifies questions answered incorrectly by the student model to determine which weaknesses require teacher guidance. Then, LEAD generates teacher rationales and similar samples to construct candidate supervision for these errors. Through supervision refinement, reliable and learnable training samples are retained from the candidate supervision. Finally, LEAD trains the student on a balanced sample mixture to correct learnable errors while preserving previously acquired knowledge. Experiments on four medical question answering benchmarks show that LEAD outperforms existing distillation methods.</abstract>
<identifier type="citekey">chen-etal-2026-identifying</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.569/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>11982</start>
<end>12004</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Identifying Student Real Weaknesses: Learnable Error-Aware Distillation for Medical Question Answering
%A Chen, Yonghe
%A Meng, Xiaoyang
%A Zhou, Xiaoding
%A Zhang, Wentian
%A Zhao, Ruimeng
%A Lian, Ying
%A Chen, Guoqiang
%A Zhang, Weiyu
%A Lu, Wenpeng
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F chen-etal-2026-identifying
%X Large language models (LLMs) show strong performance in medical question answering, but their computational costs limit deployment under resource constraints. Therefore, an increasing number of studies focus on distilling knowledge from LLMs into smaller models to improve performance in medical question answering. However, existing methods often fail to identify truly useful knowledge to distill, making it difficult to focus distillation on the student’s real weaknesses. As a result, much of the distillation budget is spent on already acquired knowledge rather than on weaknesses that still limit performance. To address this problem, we propose LEAD, a Learnable Error-Aware Distillation framework that precisely improves the weaknesses of student models. First, LEAD identifies questions answered incorrectly by the student model to determine which weaknesses require teacher guidance. Then, LEAD generates teacher rationales and similar samples to construct candidate supervision for these errors. Through supervision refinement, reliable and learnable training samples are retained from the candidate supervision. Finally, LEAD trains the student on a balanced sample mixture to correct learnable errors while preserving previously acquired knowledge. Experiments on four medical question answering benchmarks show that LEAD outperforms existing distillation methods.
%U https://aclanthology.org/2026.findings-emnlp.569/
%P 11982-12004
Markdown (Informal)
[Identifying Student Real Weaknesses: Learnable Error-Aware Distillation for Medical Question Answering](https://aclanthology.org/2026.findings-emnlp.569/) (Chen et al., Findings 2026)
ACL
- Yonghe Chen, Xiaoyang Meng, Xiaoding Zhou, Wentian Zhang, Ruimeng Zhao, Ying Lian, Guoqiang Chen, Weiyu Zhang, and Wenpeng Lu. 2026. Identifying Student Real Weaknesses: Learnable Error-Aware Distillation for Medical Question Answering. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 11982–12004, Budapest, Hungary. Association for Computational Linguistics.