@inproceedings{touchent-de-la-clergerie-2026-ontobook,
title = "{O}nto{B}ook: Ontology-Grounded Synthetic Textbooks for Medical Encoder Pretraining",
author = "Touchent, Rian and
de la Clergerie, {\'E}ric",
editor = "S{\'e}rasset, Gilles and
Gkirtzou, Katerina and
Cochez, Michael and
Kalo, Jan-Christoph",
booktitle = "Proceedings of the Knowledge Graphs and Large Language Models Workshop ({KG}-{LLM}) @ {LREC}26",
month = may,
year = "2026",
address = "Palma, Mallorca (Spain)",
publisher = "ELRA Language Resources Association (ELRA)",
url = "https://aclanthology.org/2026.kallm-1.2/",
doi = "10.63317/37ik5npnrvv6",
pages = "11--19",
abstract = "We present OntoBook, a method that converts medical ontology structure into pretraining signal for encoder language models. Our approach has three stages: random walks through ontology graphs capture hierarchical and causal relations between medical codes, a large language model reformulates these walks into fluent textbook-style prose, and the resulting text is used to train ModernCamemBERT, a 149M-parameter French encoder, with two objectives on the same data: masked language modeling and relation prediction between code pairs. On three French medical coding benchmarks (FRACCO, Cantemist-FR, Distemist-FR), OntoBook achieves significant improvements over MLM-only pretraining, with +2.5 micro-F1 on FRACCO and +8.0 micro-F1 on Distemist. We find that alignment between objectives is necessary: misaligned training, where each task uses different data, causes a 30-point degradation. We release 1.3 million LLM-reformulated medical textbooks across three French ontologies (CIM-10, CCAM, ATC) and pretrained model checkpoints."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="touchent-de-la-clergerie-2026-ontobook">
<titleInfo>
<title>OntoBook: Ontology-Grounded Synthetic Textbooks for Medical Encoder Pretraining</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rian</namePart>
<namePart type="family">Touchent</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Éric</namePart>
<namePart type="family">de la Clergerie</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Knowledge Graphs and Large Language Models Workshop (KG-LLM) @ LREC26</title>
</titleInfo>
<name type="personal">
<namePart type="given">Gilles</namePart>
<namePart type="family">Sérasset</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Katerina</namePart>
<namePart type="family">Gkirtzou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michael</namePart>
<namePart type="family">Cochez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jan-Christoph</namePart>
<namePart type="family">Kalo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Palma, Mallorca (Spain)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We present OntoBook, a method that converts medical ontology structure into pretraining signal for encoder language models. Our approach has three stages: random walks through ontology graphs capture hierarchical and causal relations between medical codes, a large language model reformulates these walks into fluent textbook-style prose, and the resulting text is used to train ModernCamemBERT, a 149M-parameter French encoder, with two objectives on the same data: masked language modeling and relation prediction between code pairs. On three French medical coding benchmarks (FRACCO, Cantemist-FR, Distemist-FR), OntoBook achieves significant improvements over MLM-only pretraining, with +2.5 micro-F1 on FRACCO and +8.0 micro-F1 on Distemist. We find that alignment between objectives is necessary: misaligned training, where each task uses different data, causes a 30-point degradation. We release 1.3 million LLM-reformulated medical textbooks across three French ontologies (CIM-10, CCAM, ATC) and pretrained model checkpoints.</abstract>
<identifier type="citekey">touchent-de-la-clergerie-2026-ontobook</identifier>
<identifier type="doi">10.63317/37ik5npnrvv6</identifier>
<location>
<url>https://aclanthology.org/2026.kallm-1.2/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>11</start>
<end>19</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T OntoBook: Ontology-Grounded Synthetic Textbooks for Medical Encoder Pretraining
%A Touchent, Rian
%A de la Clergerie, Éric
%Y Sérasset, Gilles
%Y Gkirtzou, Katerina
%Y Cochez, Michael
%Y Kalo, Jan-Christoph
%S Proceedings of the Knowledge Graphs and Large Language Models Workshop (KG-LLM) @ LREC26
%D 2026
%8 May
%I ELRA Language Resources Association (ELRA)
%C Palma, Mallorca (Spain)
%F touchent-de-la-clergerie-2026-ontobook
%X We present OntoBook, a method that converts medical ontology structure into pretraining signal for encoder language models. Our approach has three stages: random walks through ontology graphs capture hierarchical and causal relations between medical codes, a large language model reformulates these walks into fluent textbook-style prose, and the resulting text is used to train ModernCamemBERT, a 149M-parameter French encoder, with two objectives on the same data: masked language modeling and relation prediction between code pairs. On three French medical coding benchmarks (FRACCO, Cantemist-FR, Distemist-FR), OntoBook achieves significant improvements over MLM-only pretraining, with +2.5 micro-F1 on FRACCO and +8.0 micro-F1 on Distemist. We find that alignment between objectives is necessary: misaligned training, where each task uses different data, causes a 30-point degradation. We release 1.3 million LLM-reformulated medical textbooks across three French ontologies (CIM-10, CCAM, ATC) and pretrained model checkpoints.
%R 10.63317/37ik5npnrvv6
%U https://aclanthology.org/2026.kallm-1.2/
%U https://doi.org/10.63317/37ik5npnrvv6
%P 11-19
Markdown (Informal)
[OntoBook: Ontology-Grounded Synthetic Textbooks for Medical Encoder Pretraining](https://aclanthology.org/2026.kallm-1.2/) (Touchent & de la Clergerie, KaLLM 2026)
ACL