Joshua Saxe
Using the word distillation to talk about taking the textual outputs of a teacher model and using them to help train a student model reminds me of how silly I felt starting to use the term AI for the term ML
Joshua Saxe
Using the word distillation to talk about taking the textual outputs of a teacher model and using them to help train a student model reminds me of how silly I felt starting to use the term AI for the term ML