1 paper · 1 filter
Darius Catrina, Christian Bepler, Samuel Sledzieski +1
Unlike the predictable scaling laws in natural language processing and computer vision, protein language models (PLMs) scale poorly: for many tasks, models within the same family p…