TY - GEN
T1 - How can we accelerate progress towards human-like linguistic generalization?
AU - Linzen, Tal
N1 - Publisher Copyright:
© 2020 Association for Computational Linguistics
PY - 2020
Y1 - 2020
N2 - This position paper describes and critiques the Pretraining-Agnostic Identically Distributed (PAID) evaluation paradigm, which has become a central tool for measuring progress in natural language understanding. This paradigm consists of three stages: (1) pretraining of a word prediction model on a corpus of arbitrary size; (2) fine-tuning (transfer learning) on a training set representing a classification task; (3) evaluation on a test set drawn from the same distribution as that training set. This paradigm favors simple, low-bias architectures, which, first, can be scaled to process vast amounts of data, and second, can capture the fine-grained statistical properties of a particular data set, regardless of whether those properties are likely to generalize to examples of the task outside the data set. This contrasts with humans, who learn language from several orders of magnitude less data than the systems favored by this evaluation paradigm, and generalize to new tasks in a consistent way. We advocate for supplementing or replacing PAID with paradigms that reward architectures that generalize as quickly and robustly as humans.
AB - This position paper describes and critiques the Pretraining-Agnostic Identically Distributed (PAID) evaluation paradigm, which has become a central tool for measuring progress in natural language understanding. This paradigm consists of three stages: (1) pretraining of a word prediction model on a corpus of arbitrary size; (2) fine-tuning (transfer learning) on a training set representing a classification task; (3) evaluation on a test set drawn from the same distribution as that training set. This paradigm favors simple, low-bias architectures, which, first, can be scaled to process vast amounts of data, and second, can capture the fine-grained statistical properties of a particular data set, regardless of whether those properties are likely to generalize to examples of the task outside the data set. This contrasts with humans, who learn language from several orders of magnitude less data than the systems favored by this evaluation paradigm, and generalize to new tasks in a consistent way. We advocate for supplementing or replacing PAID with paradigms that reward architectures that generalize as quickly and robustly as humans.
UR - http://www.scopus.com/inward/record.url?scp=85115846851&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85115846851&partnerID=8YFLogxK
U2 - 10.18653/v1/2020.acl-main.465
DO - 10.18653/v1/2020.acl-main.465
M3 - Conference contribution
AN - SCOPUS:85115846851
T3 - Proceedings of the Annual Meeting of the Association for Computational Linguistics
SP - 5210
EP - 5217
BT - ACL 2020 - 58th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference
PB - Association for Computational Linguistics (ACL)
T2 - 58th Annual Meeting of the Association for Computational Linguistics, ACL 2020
Y2 - 5 July 2020 through 10 July 2020
ER -