Abstract
The impact of design choices on the performance of biomedical language models recently has been a subject for investigation. In this paper, we empirically study biomedical domain adaptation with large transformer models using different design choices. We evaluate the performance of our pretrained models against other existing biomedical language models in the literature. Our results show that we achieve state-of-the-art results on several biomedical domain tasks despite using similar or less computational cost compared to other models in the literature. Our findings highlight the significant effect of design choices on improving the performance of biomedical language models.- Anthology ID:
- 2021.bionlp-1.24
- Volume:
- Proceedings of the 20th Workshop on Biomedical Language Processing
- Month:
- June
- Year:
- 2021
- Address:
- Online
- Editors:
- Dina Demner-Fushman, Kevin Bretonnel Cohen, Sophia Ananiadou, Junichi Tsujii
- Venue:
- BioNLP
- SIG:
- SIGBIOMED
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 221–227
- Language:
- URL:
- https://aclanthology.org/2021.bionlp-1.24
- DOI:
- 10.18653/v1/2021.bionlp-1.24
- Cite (ACL):
- Sultan Alrowili and Vijay Shanker. 2021. BioM-Transformers: Building Large Biomedical Language Models with BERT, ALBERT and ELECTRA. In Proceedings of the 20th Workshop on Biomedical Language Processing, pages 221–227, Online. Association for Computational Linguistics.
- Cite (Informal):
- BioM-Transformers: Building Large Biomedical Language Models with BERT, ALBERT and ELECTRA (Alrowili & Shanker, BioNLP 2021)
- PDF:
- https://preview.aclanthology.org/ml4al-ingestion/2021.bionlp-1.24.pdf
- Code
- salrowili/biom-transformers
- Data
- BC5CDR, ChemProt, NCBI Disease, SQuAD