Shared computational principles for language processing in humans and deep language models

Departing from traditional linguistic models, advances in deep learning have resulted in a new type of predictive (autoregressive) deep language models (DLMs). Using a self-supervised next-word prediction task, these models generate appropriate linguistic responses in a given context. In the current...

Descripción completa

Detalles Bibliográficos
Autores principales: Goldstein, Ariel, Zada, Zaid, Buchnik, Eliav, Schain, Mariano, Price, Amy, Aubrey, Bobbi, Nastase, Samuel A., Feder, Amir, Emanuel, Dotan, Cohen, Alon, Jansen, Aren, Gazula, Harshvardhan, Choe, Gina, Rao, Aditi, Kim, Catherine, Casto, Colton, Fanda, Lora, Doyle, Werner, Friedman, Daniel, Dugan, Patricia, Melloni, Lucia, Reichart, Roi, Devore, Sasha, Flinker, Adeen, Hasenfratz, Liat, Levy, Omer, Hassidim, Avinatan, Brenner, Michael, Matias, Yossi, Norman, Kenneth A., Devinsky, Orrin, Hasson, Uri
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Nature Publishing Group US 2022
Materias:
Article
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8904253/
https://www.ncbi.nlm.nih.gov/pubmed/35260860
http://dx.doi.org/10.1038/s41593-022-01026-4

Internet

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8904253/
https://www.ncbi.nlm.nih.gov/pubmed/35260860
http://dx.doi.org/10.1038/s41593-022-01026-4

Ejemplares similares