World Models AND Tweet
Descendants partagés
1 Document
Stanford NLP Group sur Twitter : "Evidence that large pre-trained language models do successfully induce world models—because they enhance the model’s ability to predict the future)."
Tags:
2023-01-23 A propos