SciFive: a text-to-text transformer model for biomedical literature
This work addresses biomedical NLP problems for researchers and practitioners, showing incremental improvement by adapting a text-to-text transformer to the biomedical domain.
The authors tackled biomedical NLP tasks by introducing SciFive, a domain-specific T5 model pre-trained on biomedical corpora, which outperformed SOTA methods like BERT and BioBERT on tasks such as named entity relation and question-answering.
In this report, we introduce SciFive, a domain-specific T5 model that has been pre-trained on large biomedical corpora. Our model outperforms the current SOTA methods (i.e. BERT, BioBERT, Base T5) on tasks in named entity relation, relation extraction, natural language inference, and question-answering. We show that text-generation methods have significant potential in a broad array of biomedical NLP tasks, particularly those requiring longer, more complex outputs. Our results support the exploration of more difficult text generation tasks and the development of new methods in this area