Factorising

Kris Cao, Stephen Clark
2019 Proceedings of the 2019 Conference of the North  
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
more » ... abelled data. We also demonstrate that we can generate meaning-preserving syntactic paraphrases of the same AMR graph, as judged by humans.
doi:10.18653/v1/n19-1223 dblp:conf/naacl/CaoC19 fatcat:ombtvbdu3bbknobu42hwnoiodu