Structural Supervision Improves Few-Shot Learning and Syntactic Generalization in Neural Language Models

Ethan Wilcox, Peng Qian, Richard Futrell, Ryosuke Kohita, Roger Levy, Miguel Ballesteros


Abstract
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 structural supervision on learning outcomes. First, we assess few-shot learning capabilities by developing controlled experiments that probe models’ syntactic nominal number and verbal argument structure generalizations for tokens seen as few as two times during training. Second, we assess invariance properties of learned representation: the ability of a model to transfer syntactic generalizations from a base context (e.g., a simple declarative active-voice sentence) to a transformed context (e.g., an interrogative sentence). We test four models trained on the same dataset: an n-gram baseline, an LSTM, and two LSTM-variants trained with explicit structural supervision. We find that in most cases, the neural models are able to induce the proper syntactic generalizations after minimal exposure, often from just two examples during training, and that the two structurally supervised models generalize more accurately than the LSTM model. All neural models are able to leverage information learned in base contexts to drive expectations in transformed contexts, indicating that they have learned some invariance properties of syntax.
Anthology ID:
2020.emnlp-main.375
Volume:
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
Month:
November
Year:
2020
Address:
Online
Editors:
Bonnie Webber, Trevor Cohn, Yulan He, Yang Liu
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4640–4652
Language:
URL:
https://aclanthology.org/2020.emnlp-main.375
DOI:
10.18653/v1/2020.emnlp-main.375
Bibkey:
Cite (ACL):
Ethan Wilcox, Peng Qian, Richard Futrell, Ryosuke Kohita, Roger Levy, and Miguel Ballesteros. 2020. Structural Supervision Improves Few-Shot Learning and Syntactic Generalization in Neural Language Models. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 4640–4652, Online. Association for Computational Linguistics.
Cite (Informal):
Structural Supervision Improves Few-Shot Learning and Syntactic Generalization in Neural Language Models (Wilcox et al., EMNLP 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.emnlp-main.375.pdf
Video:
 https://slideslive.com/38939210
Data
CELEXPenn Treebank