Google Research is giving LLMs a "sleep" phase: the model keeps absorbing new facts and sharpens its reasoning, while limiting catastrophic forgetting and frozen knowledge cutoffs. Models that keep learning after launch. via @GoogleResearch #AIResearch
Google Research @GoogleResearchToday on the blog, learn about our new instance of a “sleep” phase in continual learning, which enables LLMs to continuously integrate new facts and enhance reasoning while mitigating catastrophic forgetting or static knowledge cutoffs. More →
00033











