Enregistré dans:
Détails bibliographiques
Auteurs principaux: Belfathi, Anas, Gallina, Ygor, Hernandez, Nicolas, Dufour, Richard, Monceaux, Laura
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2402.12036
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913243641937920
author Belfathi, Anas
Gallina, Ygor
Hernandez, Nicolas
Dufour, Richard
Monceaux, Laura
author_facet Belfathi, Anas
Gallina, Ygor
Hernandez, Nicolas
Dufour, Richard
Monceaux, Laura
contents Recent advances in pre-trained language modeling have facilitated significant progress across various natural language processing (NLP) tasks. Word masking during model training constitutes a pivotal component of language modeling in architectures like BERT. However, the prevalent method of word masking relies on random selection, potentially disregarding domain-specific linguistic attributes. In this article, we introduce an innovative masking approach leveraging genre and topicality information to tailor language models to specialized domains. Our method incorporates a ranking process that prioritizes words based on their significance, subsequently guiding the masking procedure. Experiments conducted using continual pre-training within the legal domain have underscored the efficacy of our approach on the LegalGLUE benchmark in the English language. Pre-trained language models and code are freely available for use.
format Preprint
id arxiv_https___arxiv_org_abs_2402_12036
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Language Model Adaptation to Specialized Domains through Selective Masking based on Genre and Topical Characteristics
Belfathi, Anas
Gallina, Ygor
Hernandez, Nicolas
Dufour, Richard
Monceaux, Laura
Computation and Language
Recent advances in pre-trained language modeling have facilitated significant progress across various natural language processing (NLP) tasks. Word masking during model training constitutes a pivotal component of language modeling in architectures like BERT. However, the prevalent method of word masking relies on random selection, potentially disregarding domain-specific linguistic attributes. In this article, we introduce an innovative masking approach leveraging genre and topicality information to tailor language models to specialized domains. Our method incorporates a ranking process that prioritizes words based on their significance, subsequently guiding the masking procedure. Experiments conducted using continual pre-training within the legal domain have underscored the efficacy of our approach on the LegalGLUE benchmark in the English language. Pre-trained language models and code are freely available for use.
title Language Model Adaptation to Specialized Domains through Selective Masking based on Genre and Topical Characteristics
topic Computation and Language
url https://arxiv.org/abs/2402.12036