@inproceedings{gao-etal-2026-citations,
title = "Not All Citations Are Equal:Entropy-Guided Citation Selection for Noise-Resistant Medical {LLM}",
author = "Gao, Minyu and
Xiao, Hanlin and
Wang, Ruoyu and
Yang, Shuai and
Zhang, YeXuan and
Wu, Xin and
Liu, Xingyu",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.1727/",
pages = "34599--34615",
ISBN = "979-8-89176-395-1",
abstract = "Retrieval-Augmented Generation (RAG) provides external knowledge support for large language models (LLMs) in medical applications, but retrieved contexts often contain noisy or conflicting evidence that can degrade reasoning. We observe that when internal and external knowledge disagree, models systematically prefer external citations, inadvertently injecting retrieval noise. Our analyses further show that only a subset of retrieved citations consistently improves outcomes; these effective citations exhibit markedly lower token-level entropy, linking citation entropy to model accuracy. Building on these findings, we propose a complete pipeline consisting of a training-free multi-turn reasoning framework and a post-training methodology. The training-free framework elicits internal thought, external thought, and fusion thought, and applies conflict detection and explicit denoising for complex queries. For post-training, we distill structured supervised fine-tuning (SFT) data and apply GRPO with an entropy-based citation reward that encourages selective citation of beneficial external knowledge while penalizing noisy citations. Experiments across diverse benchmarks demonstrate consistent gains in noise-resistant medical reasoning, with larger improvements on harder cases."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gao-etal-2026-citations">
<titleInfo>
<title>Not All Citations Are Equal:Entropy-Guided Citation Selection for Noise-Resistant Medical LLM</title>
</titleInfo>
<name type="personal">
<namePart type="given">Minyu</namePart>
<namePart type="family">Gao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hanlin</namePart>
<namePart type="family">Xiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruoyu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shuai</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">YeXuan</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xingyu</namePart>
<namePart type="family">Liu</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>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</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, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>Retrieval-Augmented Generation (RAG) provides external knowledge support for large language models (LLMs) in medical applications, but retrieved contexts often contain noisy or conflicting evidence that can degrade reasoning. We observe that when internal and external knowledge disagree, models systematically prefer external citations, inadvertently injecting retrieval noise. Our analyses further show that only a subset of retrieved citations consistently improves outcomes; these effective citations exhibit markedly lower token-level entropy, linking citation entropy to model accuracy. Building on these findings, we propose a complete pipeline consisting of a training-free multi-turn reasoning framework and a post-training methodology. The training-free framework elicits internal thought, external thought, and fusion thought, and applies conflict detection and explicit denoising for complex queries. For post-training, we distill structured supervised fine-tuning (SFT) data and apply GRPO with an entropy-based citation reward that encourages selective citation of beneficial external knowledge while penalizing noisy citations. Experiments across diverse benchmarks demonstrate consistent gains in noise-resistant medical reasoning, with larger improvements on harder cases.</abstract>
<identifier type="citekey">gao-etal-2026-citations</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.1727/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>34599</start>
<end>34615</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Not All Citations Are Equal:Entropy-Guided Citation Selection for Noise-Resistant Medical LLM
%A Gao, Minyu
%A Xiao, Hanlin
%A Wang, Ruoyu
%A Yang, Shuai
%A Zhang, YeXuan
%A Wu, Xin
%A Liu, Xingyu
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F gao-etal-2026-citations
%X Retrieval-Augmented Generation (RAG) provides external knowledge support for large language models (LLMs) in medical applications, but retrieved contexts often contain noisy or conflicting evidence that can degrade reasoning. We observe that when internal and external knowledge disagree, models systematically prefer external citations, inadvertently injecting retrieval noise. Our analyses further show that only a subset of retrieved citations consistently improves outcomes; these effective citations exhibit markedly lower token-level entropy, linking citation entropy to model accuracy. Building on these findings, we propose a complete pipeline consisting of a training-free multi-turn reasoning framework and a post-training methodology. The training-free framework elicits internal thought, external thought, and fusion thought, and applies conflict detection and explicit denoising for complex queries. For post-training, we distill structured supervised fine-tuning (SFT) data and apply GRPO with an entropy-based citation reward that encourages selective citation of beneficial external knowledge while penalizing noisy citations. Experiments across diverse benchmarks demonstrate consistent gains in noise-resistant medical reasoning, with larger improvements on harder cases.
%U https://aclanthology.org/2026.findings-acl.1727/
%P 34599-34615
Markdown (Informal)
[Not All Citations Are Equal:Entropy-Guided Citation Selection for Noise-Resistant Medical LLM](https://aclanthology.org/2026.findings-acl.1727/) (Gao et al., Findings 2026)
ACL
- Minyu Gao, Hanlin Xiao, Ruoyu Wang, Shuai Yang, YeXuan Zhang, Xin Wu, and Xingyu Liu. 2026. Not All Citations Are Equal:Entropy-Guided Citation Selection for Noise-Resistant Medical LLM. In Findings of the Association for Computational Linguistics: ACL 2026, pages 34599–34615, San Diego, California, United States. Association for Computational Linguistics.