Biomedicine: Stanford CRFM Introduces PubMedGPT 2.7B
Stanford CRFM (Center for Research on Foundation Models] has recently been studying domain-specific foundation models, i.e. models that are trained exclusively, or nearly exclusively, on data from a particular subdomain such as medicine or law. We believe this investigation will allow us to better understand how data composition impacts foundation models and produce more accurate and efficient systems for domain-specific downstream tasks.
We partnered with MosaicML to build PubMedGPT 2.7B, a new language model trained exclusively on biomedical abstracts and papers. This GPT-style model can achieve strong results on a variety of biomedical NLP tasks, including a new state of the art performance of 50.3% accuracy on the MedQA biomedical question answering task.
Today we are excited to make this model available to the community. As an autoregressive language model, PubMedGPT 2.7B is also capable of natural language generation. However, we have only begun to explore the generation capabilities and limitations of this model, and we emphasize that this model’s generation capabilities are only for research purposes and not suitable for production. In releasing this model, we hope to advance both the development of biomedical NLP applications and best practices for responsibly training and utilizing domain-specific language models; issues of reliability, truthfulness, and explainability are top of mind for us. We hope lessons learned from training this biomedical model can be applied to other domains such as law or finance.
PubMedGPT 2.7B was trained on all the PubMed abstracts and full documents from The Pile.
The model was trained on MosaicML Cloud, a platform designed for large workloads like LLMs. Using the Composer training library and PyTorch FSDP, it was easy to enable multi-node training across 128 A100-40GB GPUs, and the total run was completed in ~6.25 days. For more details on how we engineered and orchestrated training, see MosaicML’s companion blog post.
With a fully trained model in hand, we began exploring its capabilities in the biomedical QA space. We focused on 3 common biomedical NLP QA tasks: MedQA, PubMedQA, and BioASQ. MedQA multiple choice questions are derived from online practice exams from the USMLE, the standardized exam used to license doctors in the United States. PubMedQA and BioASQ offer up a passage and ask yes/no/maybe questions based on the passage. The PubMedQA and BioASQ contexts are derived from PubMed abstracts, with questions and answers developed by domain experts.
About Gary Price
Gary Price (email@example.com) is a librarian, writer, consultant, and frequent conference speaker based in the Washington D.C. metro area. He earned his MLIS degree from Wayne State University in Detroit. Price has won several awards including the SLA Innovations in Technology Award and Alumnus of the Year from the Wayne St. University Library and Information Science Program. From 2006-2009 he was Director of Online Information Services at Ask.com. Gary is also the co-founder of infoDJ an innovation research consultancy supporting corporate product and business model teams with just-in-time fact and insight finding.