There's a LOT of interesting stuff happening with large language models in biology: Transformer trained on human genomes, learns to identify different genomic

There's a LOT of interesting stuff happening with large language models in biology: Transformer trained on human genomes, learns to identify different genomic elements like enhancers without knowing what enhancers are https://www.biorxiv.org/content/10.1101/2023.01.11.523679v2 https://www.biorxiv.org/content/10.1101/2023.01.11.523679v2 Similarly, a transformer trained on only plant genomes 'knows' how the strength of genomic variants' impact https://www.biorxiv.org/content/10.1101/2022.08.22.504706v2 https://www.biorxiv.org/content/10.1101/2022.08.22.504706v2 Lots of experiments happening around GPT-4 too; Using regular GPT-4 to curate cell type annotations https://www.biorxiv.org/content/10.1101/2023.04.16.537094v1 https://www.biorxiv.org/content/10.1101/2023.04.16.537094v1 Testing GPT-4 across a variety of 'standard' tasks like linking gene IDs to protein IDs https://www.biorxiv.org/content/10.1101/2023.03.11.532238v1 https://www.biorxiv.org/content/10.1101/2023.03.11.532238v1