Title :

DrBenchmark: A Large Language Understanding Evaluation Benchmark for French Biomedical Domain

Author(s) :

Labrak, Yanis [Auteur]
Laboratoire Informatique d'Avignon [LIA]
Bazoge, Adrien [Auteur]
Traitement Automatique du Langage Naturel [LS2N - équipe TALN ]
El Khettari, Oumaima [Auteur]
Traitement Automatique du Langage Naturel [LS2N - équipe TALN ]
Rouvier, Mickaël [Auteur]
Laboratoire Informatique d'Avignon [LIA]
Constant Dit Beaufils, Pacôme [Auteur]
Centre Hospitalier Universitaire de Nantes = Nantes University Hospital [CHU Nantes]
Grabar, Natalia [Auteur]
Savoirs, Textes, Langage (STL) - UMR 8163 [STL]
Daille, Béatrice [Auteur]
Traitement Automatique du Langage Naturel [LS2N - équipe TALN ]
Quiniou, Solen [Auteur]
Traitement Automatique du Langage Naturel [LS2N - équipe TALN ]
Morin, Emmanuel [Auteur]
Traitement Automatique du Langage Naturel [LS2N - équipe TALN ]
Gourraud, Pierre‐antoine [Auteur]
Dufour, Richard [Auteur]
Traitement Automatique du Langage Naturel [LS2N - équipe TALN ]
Laboratoire Informatique d'Avignon [LIA]

Conference title :

Fourteenth Language Resources and Evaluation Conference (LREC-COLING 2024)

Conference organizers(s) :

Nicoletta Calzolari
Min-Yen Kan

City :

Torino

Country :

Italie

Start date of the conference :

2024-05-20

English keyword(s) :

NLP evaluation
Benchmarking
Medical domain
French language
Transformers

HAL domain(s) :

Informatique [cs]
Informatique [cs]/Intelligence artificielle [cs.AI]
Informatique [cs]/Informatique et langage [cs.CL]
Informatique [cs]/Apprentissage [cs.LG]
Informatique [cs]/Traitement du texte et du document

English abstract : [en]

The biomedical domain has sparked a significant interest in the field of Natural Language Processing (NLP), which has seen substantial advancements with pre-trained language models (PLMs). However, comparing these models ...
Show more >The biomedical domain has sparked a significant interest in the field of Natural Language Processing (NLP), which has seen substantial advancements with pre-trained language models (PLMs). However, comparing these models has proven challenging due to variations in evaluation protocols across different models. A fair solution is to aggregate diverse downstream tasks into a benchmark, allowing for the assessment of intrinsic PLMs qualities from various perspectives. Although still limited to few languages, this initiative has been undertaken in the biomedical field, notably English and Chinese. This limitation hampers the evaluation of the latest French biomedical models, as they are either assessed on a minimal number of tasks with non-standardized protocols or evaluated using general downstream tasks. To bridge this research gap and account for the unique sensitivities of French, we present the first-ever publicly available French biomedical language understanding benchmark called DrBenchmark. It encompasses 20 diversified tasks, including named-entity recognition, part-of-speech tagging, question-answering, semantic textual similarity, and classification. We evaluate 8 state-of-the-art pre-trained masked language models (MLMs) on general and biomedical-specific data, as well as English specific MLMs to assess their cross-lingual capabilities. Our experiments reveal that no single model excels across all tasks, while generalist models are sometimes still competitive.Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

ANR Project :

Grands modèles de langue adaptables et souverains pour le domaine médical français
AI by / for Human, Health and Industry

Collections :

• Savoirs, Textes, Langage (STL) - UMR 8163

Source :

Harvested from HAL