papers AI Learner
The Github is limit! Click to go to the new site.

Factorising AMR generation through syntax

2019-04-03
Kris Cao, Stephen Clark

Abstract

Generating from Abstract Meaning Representation (AMR) is an underspecified problem, as many syntactic decisions are not constrained by the semantic graph. To explicitly account for this underspecification, 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.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1804.07707

PDF

http://arxiv.org/pdf/1804.07707


Comments

Content