Finetuning LLaMA on Medical Papers

The result is PMC-LLaMA trained on 4.8 million Medical papers.

Photo by National Cancer Institute on Unsplash

Welcome Back!

As some of you know I’m a big followers of A.I. in the future of healthcare. So I thought I had to cover this. From this tweet.

Finetuning LLaMA on Medical Papers

- fine-tunes LLaMA on 4.8 million biomedical papers - enhances capabilities in the medical domain - the proposed model, PMC-LLaMA, achieves high performance on biomedical QA benchmarks

paper: https://arxiv.org/abs/2304.14454

code: https://github.com/chaoyi-wu/PMC-LLaMA

With the emergence of a “Linux of A.I.” (mid 2023) moment this is more interesting than it at first appears.