Factorising AMR generation through syntax

Kris Cao, Stephen Clark


Abstract
Generating from Abstract Meaning Representation (AMR) is an underspecified problem, as many syntactic decisions are not specified by the semantic graph. To explicitly account for this variation, we break down generating from AMR into two steps: first generate a syntactic structure, and then generate the surface form. We show that decomposing the generation process this way leads to state-of-the-art single model performance generating from AMR without additional unlabelled data. We also demonstrate that we can generate meaning-preserving syntactic paraphrases of the same AMR graph, as judged by humans.
Anthology ID:
N19-1223
Volume:
Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)
Month:
June
Year:
2019
Address:
Minneapolis, Minnesota
Editors:
Jill Burstein, Christy Doran, Thamar Solorio
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2157–2163
Language:
URL:
https://aclanthology.org/N19-1223
DOI:
10.18653/v1/N19-1223
Bibkey:
Cite (ACL):
Kris Cao and Stephen Clark. 2019. Factorising AMR generation through syntax. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 2157–2163, Minneapolis, Minnesota. Association for Computational Linguistics.
Cite (Informal):
Factorising AMR generation through syntax (Cao & Clark, NAACL 2019)
Copy Citation:
PDF:
https://aclanthology.org/N19-1223.pdf
Video:
 https://aclanthology.org/N19-1223.mp4