About This Document
- sl:arxiv_author :
- sl:arxiv_firstAuthor : Yu Gu
- sl:arxiv_num : 2007.15779
- sl:arxiv_published : 2020-07-31T00:04:15Z
- sl:arxiv_summary : Pretraining large neural language models, such as BERT, has led to impressive
gains on many natural language processing (NLP) tasks. However, most
pretraining efforts focus on general domain corpora, such as newswire and Web.
A prevailing assumption is that even domain-specific pretraining can benefit by
starting from general-domain language models. In this paper, we challenge this
assumption by showing that for domains with abundant unlabeled text, such as
biomedicine, pretraining language models from scratch results in substantial
gains over continual pretraining of general-domain language models. To
facilitate this investigation, we compile a comprehensive biomedical NLP
benchmark from publicly-available datasets. Our experiments show that
domain-specific pretraining serves as a solid foundation for a wide range of
biomedical NLP tasks, leading to new state-of-the-art results across the board.
Further, in conducting a thorough evaluation of modeling choices, both for
pretraining and task-specific fine-tuning, we discover that some common
practices are unnecessary with BERT models, such as using complex tagging
schemes in named entity recognition (NER). To help accelerate research in
biomedical NLP, we have released our state-of-the-art pretrained and
task-specific models for the community, and created a leaderboard featuring our
BLURB benchmark (short for Biomedical Language Understanding & Reasoning
Benchmark) at https://aka.ms/BLURB.@en
- sl:arxiv_title : Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing@en
- sl:arxiv_updated : 2021-02-11T19:13:59Z
- sl:bookmarkOf : https://arxiv.org/abs/2007.15779
- sl:creationDate : 2021-04-11
- sl:creationTime : 2021-04-11T16:38:59Z
Documents with similar tags (experimental)