1 paper · 1 filter
Junseo Hwang, Wonguk Cho, Taesup Kim
Fine-tuning large foundation models is essential for building expert models tailored to specialized tasks and domains, but fully updating billions of parameters is computationally…