
Sangyun Lee
@sang_yun_lee • 1,512 subscribers
PhD student @CMU_ECE | ex-intern @nvidia | Generative models
Videos

Almost all animals sleep. Why don’t LMs? Introducing our new work on language model sleep. tl;dr : A periodic, recurrent “sleep” phase allows LMs to digest their context and transfer it into their weights, improving recall and reasoning on challenging tasks.
Sangyun Lee124,820 Aufrufe • vor 2 Monaten
Keine weiteren Inhalte verfügbar