@inproceedings{qiu-etal-2026-r3,
title = "{R}$^3$ -{MDT}: A Retain-Rerank-Revise Framework Based on Uncertainty and Consistency for Medical Decision Tree Extraction",
author = "Qiu, Boyu and
Weng, Xiaolong and
Wang, Zehua and
Tang, Buzhou and
Xiong, Ying and
Zhang, Min",
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.1757/",
pages = "37577--37594",
ISBN = "979-8-89176-462-0",
abstract = "Automatic extraction of Medical Decision Trees (MDTs) is an important task for Clinical Decision Support Systems, but remains challenging due to the complex logical structures involved. Recent approaches based on Large Language Models (LLMs) commonly adopt generate-then-revise pipelines with self-correction to improve reasoning accuracy. However, we find that applying revision uniformly across all samples can be harmful, as it often alters already correct and stable predictions, leading to over-correction on simple cases. To alleviate this issue, we propose R$^3$-MDT, an adaptive inference framework that dynamically selects reasoning strategies based on sample difficulty. We use model uncertainty and output consistency as internal signals to estimate difficulty and route samples into Easy, Medium, and Hard categories. For Easy samples with confident and consistent predictions, we directly retain the initial outputs. Medium-difficulty samples, which exhibit ambiguity among multiple plausible candidates, are handled via reranking to select the optimal decision tree. Hard samples, characterized by high uncertainty and structural inconsistency, are revised to rectify intractable generation failures. By adaptively selecting inference strategies based on difficulty levels, R$^3$-MDT avoids unnecessary revisions while effectively correcting underlying reasoning failures. Experiments on the Text2DT benchmark show that R$^3$-MDT alleviates the over-correction problem and achieves state-of-the-art performance, while auxiliary language- and task-shift evaluations indicate the broader transferability of its routing principle."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="qiu-etal-2026-r3">
<titleInfo>
<title>R³ -MDT: A Retain-Rerank-Revise Framework Based on Uncertainty and Consistency for Medical Decision Tree Extraction</title>
</titleInfo>
<name type="personal">
<namePart type="given">Boyu</namePart>
<namePart type="family">Qiu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaolong</namePart>
<namePart type="family">Weng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zehua</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Buzhou</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ying</namePart>
<namePart type="family">Xiong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Min</namePart>
<namePart type="family">Zhang</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>Automatic extraction of Medical Decision Trees (MDTs) is an important task for Clinical Decision Support Systems, but remains challenging due to the complex logical structures involved. Recent approaches based on Large Language Models (LLMs) commonly adopt generate-then-revise pipelines with self-correction to improve reasoning accuracy. However, we find that applying revision uniformly across all samples can be harmful, as it often alters already correct and stable predictions, leading to over-correction on simple cases. To alleviate this issue, we propose R³-MDT, an adaptive inference framework that dynamically selects reasoning strategies based on sample difficulty. We use model uncertainty and output consistency as internal signals to estimate difficulty and route samples into Easy, Medium, and Hard categories. For Easy samples with confident and consistent predictions, we directly retain the initial outputs. Medium-difficulty samples, which exhibit ambiguity among multiple plausible candidates, are handled via reranking to select the optimal decision tree. Hard samples, characterized by high uncertainty and structural inconsistency, are revised to rectify intractable generation failures. By adaptively selecting inference strategies based on difficulty levels, R³-MDT avoids unnecessary revisions while effectively correcting underlying reasoning failures. Experiments on the Text2DT benchmark show that R³-MDT alleviates the over-correction problem and achieves state-of-the-art performance, while auxiliary language- and task-shift evaluations indicate the broader transferability of its routing principle.</abstract>
<identifier type="citekey">qiu-etal-2026-r3</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1757/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>37577</start>
<end>37594</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T R³ -MDT: A Retain-Rerank-Revise Framework Based on Uncertainty and Consistency for Medical Decision Tree Extraction
%A Qiu, Boyu
%A Weng, Xiaolong
%A Wang, Zehua
%A Tang, Buzhou
%A Xiong, Ying
%A Zhang, Min
%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 qiu-etal-2026-r3
%X Automatic extraction of Medical Decision Trees (MDTs) is an important task for Clinical Decision Support Systems, but remains challenging due to the complex logical structures involved. Recent approaches based on Large Language Models (LLMs) commonly adopt generate-then-revise pipelines with self-correction to improve reasoning accuracy. However, we find that applying revision uniformly across all samples can be harmful, as it often alters already correct and stable predictions, leading to over-correction on simple cases. To alleviate this issue, we propose R³-MDT, an adaptive inference framework that dynamically selects reasoning strategies based on sample difficulty. We use model uncertainty and output consistency as internal signals to estimate difficulty and route samples into Easy, Medium, and Hard categories. For Easy samples with confident and consistent predictions, we directly retain the initial outputs. Medium-difficulty samples, which exhibit ambiguity among multiple plausible candidates, are handled via reranking to select the optimal decision tree. Hard samples, characterized by high uncertainty and structural inconsistency, are revised to rectify intractable generation failures. By adaptively selecting inference strategies based on difficulty levels, R³-MDT avoids unnecessary revisions while effectively correcting underlying reasoning failures. Experiments on the Text2DT benchmark show that R³-MDT alleviates the over-correction problem and achieves state-of-the-art performance, while auxiliary language- and task-shift evaluations indicate the broader transferability of its routing principle.
%U https://aclanthology.org/2026.findings-emnlp.1757/
%P 37577-37594
Markdown (Informal)
[R3 -MDT: A Retain-Rerank-Revise Framework Based on Uncertainty and Consistency for Medical Decision Tree Extraction](https://aclanthology.org/2026.findings-emnlp.1757/) (Qiu et al., Findings 2026)
ACL