AMR Dependency Parsing with a Typed Semantic Algebra
Computation and Language
2021-06-10 v1
Abstract
We present a semantic parser for Abstract Meaning Representations which learns to parse strings into tree representations of the compositional structure of an AMR graph. This allows us to use standard neural techniques for supertagging and dependency tree parsing, constrained by a linguistically principled type system. We present two approximative decoding algorithms, which achieve state-of-the-art accuracy and outperform strong baselines.
Cite
@article{arxiv.1805.11465,
title = {AMR Dependency Parsing with a Typed Semantic Algebra},
author = {Jonas Groschwitz and Matthias Lindemann and Meaghan Fowlie and Mark Johnson and Alexander Koller},
journal= {arXiv preprint arXiv:1805.11465},
year = {2021}
}
Comments
This paper will be presented at ACL 2018 (see https://acl2018.org/programme/papers/)