@inproceedings{he-etal-2026-medfact,
title = "{M}ed{F}act: Benchmarking the Fact-Checking Capabilities of Large Language Models on {C}hinese Medical Texts",
author = "He, Jiayi and
Huang, Yangmin and
Du, Qianyun and
Zhou, Xiangying and
He, Zhiyang and
Hu, Jiaxue and
Tao, Xiaodong and
Lai, Lixian",
editor = "Mille, Simon and
Gehrmann, Sebastian and
Schmidtov{\'a}, Patr{\'i}cia and
Du{\v{s}}ek, Ond{\v{r}}ej and
Fadaee, Marzieh and
Lo, Kyle and
Santus, Enrico and
Stanovsky, Gabriel",
booktitle = "Proceedings of the Fifth Workshop on Generation, Evaluation and Metrics ({GEM})",
month = jul,
year = "2026",
address = "San Diego, California, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.gem-main.59/",
pages = "604--652",
ISBN = "979-8-89176-423-1",
abstract = "Deploying Large Language Models (LLMs) in medical applications requires rigorous fact-checking to ensure patient safety and regulatory compliance. We introduce **MedFact**, a challenging Chinese medical fact-checking benchmark with 2,116 expert-annotated instances from diverse real-world texts, spanning 13 specialties, 8 error types, 4 writing styles, and 5 difficulty levels. Construction uses a hybrid AI-human framework where iterative expert feedback refines AI-driven, multi-criteria filtering to ensure high quality and difficulty. We evaluate 20 leading LLMs on veracity classification and error localization, and results show that models can often determine whether text contains errors but struggle to localize them precisely, with top performers falling short of human performance. Our analysis reveals an ``over-criticism'' phenomenon, where models misidentify correct information as erroneous, a tendency that is aggravated by advanced reasoning techniques such as multi-agent collaboration and inference-time scaling. MedFact highlights the challenges of deploying medical LLMs and provides resources to develop factually reliable medical AI systems."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="he-etal-2026-medfact">
<titleInfo>
<title>MedFact: Benchmarking the Fact-Checking Capabilities of Large Language Models on Chinese Medical Texts</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jiayi</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yangmin</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qianyun</namePart>
<namePart type="family">Du</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiangying</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhiyang</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiaxue</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaodong</namePart>
<namePart type="family">Tao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lixian</namePart>
<namePart type="family">Lai</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>Proceedings of the Fifth Workshop on Generation, Evaluation and Metrics (GEM)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Mille</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sebastian</namePart>
<namePart type="family">Gehrmann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Patrícia</namePart>
<namePart type="family">Schmidtová</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ondřej</namePart>
<namePart type="family">Dušek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marzieh</namePart>
<namePart type="family">Fadaee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kyle</namePart>
<namePart type="family">Lo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Enrico</namePart>
<namePart type="family">Santus</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gabriel</namePart>
<namePart type="family">Stanovsky</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, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-423-1</identifier>
</relatedItem>
<abstract>Deploying Large Language Models (LLMs) in medical applications requires rigorous fact-checking to ensure patient safety and regulatory compliance. We introduce **MedFact**, a challenging Chinese medical fact-checking benchmark with 2,116 expert-annotated instances from diverse real-world texts, spanning 13 specialties, 8 error types, 4 writing styles, and 5 difficulty levels. Construction uses a hybrid AI-human framework where iterative expert feedback refines AI-driven, multi-criteria filtering to ensure high quality and difficulty. We evaluate 20 leading LLMs on veracity classification and error localization, and results show that models can often determine whether text contains errors but struggle to localize them precisely, with top performers falling short of human performance. Our analysis reveals an “over-criticism” phenomenon, where models misidentify correct information as erroneous, a tendency that is aggravated by advanced reasoning techniques such as multi-agent collaboration and inference-time scaling. MedFact highlights the challenges of deploying medical LLMs and provides resources to develop factually reliable medical AI systems.</abstract>
<identifier type="citekey">he-etal-2026-medfact</identifier>
<location>
<url>https://aclanthology.org/2026.gem-main.59/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>604</start>
<end>652</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T MedFact: Benchmarking the Fact-Checking Capabilities of Large Language Models on Chinese Medical Texts
%A He, Jiayi
%A Huang, Yangmin
%A Du, Qianyun
%A Zhou, Xiangying
%A He, Zhiyang
%A Hu, Jiaxue
%A Tao, Xiaodong
%A Lai, Lixian
%Y Mille, Simon
%Y Gehrmann, Sebastian
%Y Schmidtová, Patrícia
%Y Dušek, Ondřej
%Y Fadaee, Marzieh
%Y Lo, Kyle
%Y Santus, Enrico
%Y Stanovsky, Gabriel
%S Proceedings of the Fifth Workshop on Generation, Evaluation and Metrics (GEM)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, USA
%@ 979-8-89176-423-1
%F he-etal-2026-medfact
%X Deploying Large Language Models (LLMs) in medical applications requires rigorous fact-checking to ensure patient safety and regulatory compliance. We introduce **MedFact**, a challenging Chinese medical fact-checking benchmark with 2,116 expert-annotated instances from diverse real-world texts, spanning 13 specialties, 8 error types, 4 writing styles, and 5 difficulty levels. Construction uses a hybrid AI-human framework where iterative expert feedback refines AI-driven, multi-criteria filtering to ensure high quality and difficulty. We evaluate 20 leading LLMs on veracity classification and error localization, and results show that models can often determine whether text contains errors but struggle to localize them precisely, with top performers falling short of human performance. Our analysis reveals an “over-criticism” phenomenon, where models misidentify correct information as erroneous, a tendency that is aggravated by advanced reasoning techniques such as multi-agent collaboration and inference-time scaling. MedFact highlights the challenges of deploying medical LLMs and provides resources to develop factually reliable medical AI systems.
%U https://aclanthology.org/2026.gem-main.59/
%P 604-652
Markdown (Informal)
[MedFact: Benchmarking the Fact-Checking Capabilities of Large Language Models on Chinese Medical Texts](https://aclanthology.org/2026.gem-main.59/) (He et al., GEM 2026)
ACL
- Jiayi He, Yangmin Huang, Qianyun Du, Xiangying Zhou, Zhiyang He, Jiaxue Hu, Xiaodong Tao, and Lixian Lai. 2026. MedFact: Benchmarking the Fact-Checking Capabilities of Large Language Models on Chinese Medical Texts. In Proceedings of the Fifth Workshop on Generation, Evaluation and Metrics (GEM), pages 604–652, San Diego, California, USA. Association for Computational Linguistics.