Learning Syntax from Naturally-Occurring Bracketings
Tianze Shi, Ozan İrsoy, Igor Malioutov, Lillian Lee
Abstract
Naturally-occurring bracketings, such as answer fragments to natural language questions and hyperlinks on webpages, can reflect human syntactic intuition regarding phrasal boundaries. Their availability and approximate correspondence to syntax make them appealing as distant information sources to incorporate into unsupervised constituency parsing. But they are noisy and incomplete; to address this challenge, we develop a partial-brackets-aware structured ramp loss in learning. Experiments demonstrate that our distantly-supervised models trained on naturally-occurring bracketing data are more accurate in inducing syntactic structures than competing unsupervised systems. On the English WSJ corpus, our models achieve an unlabeled F1 score of 68.9 for constituency parsing.
BibTeX
@inproceedings{shi-etal-2021-learning,
title = "Learning Syntax from Naturally-Occurring Bracketings",
author = "Shi, Tianze and
{\.I}rsoy, Ozan and
Malioutov, Igor and
Lee, Lillian",
editor = "Toutanova, Kristina and
Rumshisky, Anna and
Zettlemoyer, Luke and
Hakkani-Tur, Dilek and
Beltagy, Iz and
Bethard, Steven and
Cotterell, Ryan and
Chakraborty, Tanmoy and
Zhou, Yichao",
booktitle = "Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies",
month = jun,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.naacl-main.234/",
doi = "10.18653/v1/2021.naacl-main.234",
pages = "2941--2949"
}