Structural Supervision Improves Few-Shot Learning and Syntactic Generalization in Neural Language Models
Humans can learn structural properties about a word from minimal experience, and deploy their learned syntactic representations uniformly in different grammatical contexts. We assess the ability of modern neural language models to reproduce this behavior in English and evaluate the effect of structu...
Main Authors: | Wilcox, Ethan (Author), Qian, Peng (Author), Futrell, Richard (Author), Kohita, Ryosuke (Author), Levy, Roger (Author), Ballesteros, Miguel (Author) |
---|---|
Format: | Article |
Language: | English |
Published: |
Association for Computational Linguistics (ACL),
2021-12-01T17:46:55Z.
|
Subjects: | |
Online Access: | Get fulltext |
Similar Items
-
Structural Supervision Improves Learning of Non-Local Grammatical Dependencies
by: Wilcox, Ethan, et al.
Published: (2022) -
A Systematic Assessment of Syntactic Generalization in Neural Language Models
by: Hu, Jennifer, et al.
Published: (2021) -
Hierarchical Representation in Neural Language Models: Suppression and Recovery of Expectations
by: Wilcox, Ethan, et al.
Published: (2021) -
Neural language models as psycholinguistic subjects: Representations of syntactic state
by: Qian, Peng, et al.
Published: (2021) -
Augmenting Few-Shot Learning With Supervised Contrastive Learning
by: Taemin Lee, et al.
Published: (2021-01-01)