A Unifying Theory of Transition-based and Sequence Labeling Parsing
Computation and Language
2020-11-03 v1 Formal Languages and Automata Theory
Abstract
We define a mapping from transition-based parsing algorithms that read sentences from left to right to sequence labeling encodings of syntactic trees. This not only establishes a theoretical relation between transition-based parsing and sequence-labeling parsing, but also provides a method to obtain new encodings for fast and simple sequence labeling parsing from the many existing transition-based parsers for different formalisms. Applying it to dependency parsing, we implement sequence labeling versions of four algorithms, showing that they are learnable and obtain comparable performance to existing encodings.
Cite
@article{arxiv.2011.00584,
title = {A Unifying Theory of Transition-based and Sequence Labeling Parsing},
author = {Carlos Gómez-Rodríguez and Michalina Strzyz and David Vilares},
journal= {arXiv preprint arXiv:2011.00584},
year = {2020}
}
Comments
Camera-ready version (final peer-reviewed manuscript) to appear at proceedings of COLING 2020. 18 pages (incl. appendices)