Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing

🤖 Plain-English Summary

Pretraining large neural language models, such as BERT, has led to impressive gains on many natural language processing (NLP) tasks. 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.

🔑 Key Findings

• 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 article, 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.

🔗 Related Resources