The human knowledge loophole in the 'bitter lesson' for LLMs

摘要

Are LLMs a proof that the 'bitter lesson' holds for NLP? Perhaps the opposite is true: they work due to the scale of human data, and not just computation.