@inproceedings{lamarche-etal-2026-medhal,
title = "{M}ed{H}al: A Synthetic Dataset for Medical Hallucination Detection",
author = "Lamarche, Fabrice and
Mehenni, Gaya and
Fard, Neshat Elhami and
Rios-Ibacache, Odette and
Wang, Li Ming and
Kildea, John and
Zouaq, Amal",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the {A}sia-Pacific Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.aacl-long.208/",
pages = "4599--4619",
ISBN = "979-8-89176-439-2",
abstract = "Hallucination, the generation of non factual content by AI systems, poses a serious risk in medical contexts, where errors can directly affect patient outcomes. We present MedHal, a novel large-scale dataset specifically designed to assess capabilities and train models on the task of hallucination detection in medical texts. Current hallucination detection methods face significant limitations when applied to specialized domains like medicine, where they can have disastrous consequences. MedHal addresses this issue by incorporating diverse medical text sources and tasks covering both intrinsic and extrinsic hallucinations, and by providing a substantial volume of data samples suitable for training medical hallucination detection models. We demonstrate MedHal{'}s utility by training and evaluating a baseline medical hallucination detection model, showing improvements over general-purpose hallucination detection approaches. This resource enables more efficient evaluation and training of medical text generation systems while reducing reliance on costly expert review, potentially accelerating the development of medical AI research."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lamarche-etal-2026-medhal">
<titleInfo>
<title>MedHal: A Synthetic Dataset for Medical Hallucination Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Fabrice</namePart>
<namePart type="family">Lamarche</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gaya</namePart>
<namePart type="family">Mehenni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Neshat</namePart>
<namePart type="given">Elhami</namePart>
<namePart type="family">Fard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Odette</namePart>
<namePart type="family">Rios-Ibacache</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Li</namePart>
<namePart type="given">Ming</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">John</namePart>
<namePart type="family">Kildea</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Amal</namePart>
<namePart type="family">Zouaq</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-439-2</identifier>
</relatedItem>
<abstract>Hallucination, the generation of non factual content by AI systems, poses a serious risk in medical contexts, where errors can directly affect patient outcomes. We present MedHal, a novel large-scale dataset specifically designed to assess capabilities and train models on the task of hallucination detection in medical texts. Current hallucination detection methods face significant limitations when applied to specialized domains like medicine, where they can have disastrous consequences. MedHal addresses this issue by incorporating diverse medical text sources and tasks covering both intrinsic and extrinsic hallucinations, and by providing a substantial volume of data samples suitable for training medical hallucination detection models. We demonstrate MedHal’s utility by training and evaluating a baseline medical hallucination detection model, showing improvements over general-purpose hallucination detection approaches. This resource enables more efficient evaluation and training of medical text generation systems while reducing reliance on costly expert review, potentially accelerating the development of medical AI research.</abstract>
<identifier type="citekey">lamarche-etal-2026-medhal</identifier>
<location>
<url>https://aclanthology.org/2026.aacl-long.208/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>4599</start>
<end>4619</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T MedHal: A Synthetic Dataset for Medical Hallucination Detection
%A Lamarche, Fabrice
%A Mehenni, Gaya
%A Fard, Neshat Elhami
%A Rios-Ibacache, Odette
%A Wang, Li Ming
%A Kildea, John
%A Zouaq, Amal
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-439-2
%F lamarche-etal-2026-medhal
%X Hallucination, the generation of non factual content by AI systems, poses a serious risk in medical contexts, where errors can directly affect patient outcomes. We present MedHal, a novel large-scale dataset specifically designed to assess capabilities and train models on the task of hallucination detection in medical texts. Current hallucination detection methods face significant limitations when applied to specialized domains like medicine, where they can have disastrous consequences. MedHal addresses this issue by incorporating diverse medical text sources and tasks covering both intrinsic and extrinsic hallucinations, and by providing a substantial volume of data samples suitable for training medical hallucination detection models. We demonstrate MedHal’s utility by training and evaluating a baseline medical hallucination detection model, showing improvements over general-purpose hallucination detection approaches. This resource enables more efficient evaluation and training of medical text generation systems while reducing reliance on costly expert review, potentially accelerating the development of medical AI research.
%U https://aclanthology.org/2026.aacl-long.208/
%P 4599-4619
Markdown (Informal)
[MedHal: A Synthetic Dataset for Medical Hallucination Detection](https://aclanthology.org/2026.aacl-long.208/) (Lamarche et al., AACL-IJCNLP 2026)
ACL
- Fabrice Lamarche, Gaya Mehenni, Neshat Elhami Fard, Odette Rios-Ibacache, Li Ming Wang, John Kildea, and Amal Zouaq. 2026. MedHal: A Synthetic Dataset for Medical Hallucination Detection. In Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4599–4619, Hengqin, China. Association for Computational Linguistics.