Zobrazeno 1 - 10
of 96
pro vyhledávání: '"William, Merrill"'
Publikováno v:
Transactions of the Association for Computational Linguistics, Vol 11, Pp 617-634 (2023)
AbstractMany current NLP systems are built from language models trained to optimize unsupervised objectives on large amounts of raw text. Under what conditions might such a procedure acquire meaning? Our systematic experiments with synthetic data rev
Externí odkaz:
https://doaj.org/article/476b5a06e2e84082827eac4be7eaf3e8
Autor:
William Merrill, Ashish Sabharwal
Publikováno v:
Transactions of the Association for Computational Linguistics, Vol 11, Pp 531-545 (2023)
AbstractDespite their omnipresence in modern NLP, characterizing the computational power of transformer neural nets remains an interesting open question. We prove that transformers whose arithmetic precision is logarithmic in the number of input toke
Externí odkaz:
https://doaj.org/article/29f71320a4e74b8c80c7db0491d6dd2c
Autor:
William Merrill Decker
An compelling coming-of-age memoir that presents a portrait of suburban life in upstate New York shaped by the Civil Rights Movement, Vietnam and the constant threat of Nuclear exchange during the 1950's/early 1960's.
Publikováno v:
Transactions of the Association for Computational Linguistics, Vol 9, Pp 1047-1060 (2021)
AbstractLanguage models trained on billions of tokens have recently led to unprecedented results on many NLP tasks. This success raises the question of whether, in principle, a system can ever “understand” raw text without access to some form of
Externí odkaz:
https://doaj.org/article/01b20bd81c594f82968f4a363577d1ef
Autor:
William Merrill
Publikováno v:
Developments in Language Theory ISBN: 9783031332630
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::11bfdf9802539763811083b6ac326907
https://doi.org/10.1007/978-3-031-33264-7_1
https://doi.org/10.1007/978-3-031-33264-7_1
Publikováno v:
Transactions of the Association for Computational Linguistics. 9:1047-1060
Language models trained on billions of tokens have recently led to unprecedented results on many NLP tasks. This success raises the question of whether, in principle, a system can ever “understand” raw text without access to some form of groundin
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::0f98a29d92138feb84480ce8d26dd0d6
https://doi.org/10.1162/tacl_a_00412
https://doi.org/10.1162/tacl_a_00412
Autor:
William Merrill Decker
Publikováno v:
Reviews in American History. 49:250-258
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::1beabe3becc249cd97386a9b35f943e2
https://doi.org/10.1353/rah.2021.0026
https://doi.org/10.1353/rah.2021.0026