Learning Contextually Informed Representations for Linear-Time Discourse Parsing

Yang Liu, Mirella Lapata


Abstract
Recent advances in RST discourse parsing have focused on two modeling paradigms: (a) high order parsers which jointly predict the tree structure of the discourse and the relations it encodes; or (b) linear-time parsers which are efficient but mostly based on local features. In this work, we propose a linear-time parser with a novel way of representing discourse constituents based on neural networks which takes into account global contextual information and is able to capture long-distance dependencies. Experimental results show that our parser obtains state-of-the art performance on benchmark datasets, while being efficient (with time complexity linear in the number of sentences in the document) and requiring minimal feature engineering.
Anthology ID:
D17-1133
Volume:
Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing
Month:
September
Year:
2017
Address:
Copenhagen, Denmark
Venue:
EMNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
1289–1298
Language:
URL:
https://aclanthology.org/D17-1133
DOI:
10.18653/v1/D17-1133
Bibkey:
Cite (ACL):
Yang Liu and Mirella Lapata. 2017. Learning Contextually Informed Representations for Linear-Time Discourse Parsing. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 1289–1298, Copenhagen, Denmark. Association for Computational Linguistics.
Cite (Informal):
Learning Contextually Informed Representations for Linear-Time Discourse Parsing (Liu & Lapata, EMNLP 2017)
Copy Citation:
PDF:
https://preview.aclanthology.org/remove-xml-comments/D17-1133.pdf