3 ms·
You must be confused. This research is about reinforcement learning, not about large language models.
by smokel 2y ago
You must be confused. This research is about reinforcement learning, not about large language models.
- ninetyninenine 2y agoIt's parroting human reinforcement.
- danijar 2y agoIt actually has no human data as input and learns by itself in the environment, that's the point of the accomplishment! :)
- ninetyninenine 2y agoThat's what humans do right? So it's parroting us.