We describe a baseline dependency parsing system for the CoNLL2017 Shared Task. This system, which we call "ParseySaurus," uses the DRAGNN framework [Kong et al, 2017] to combine transition-based recurrent parsing and tagging with character-based word representations. On the v1.3 Universal Dependencies Treebanks, the new system outpeforms the publicly available, state-of-the-art "Parsey's Cousins" models by 3.47% absolute Labeled Accuracy Score (LAS) across 52 treebanks.
@article{arxiv.1703.04929,
title = {SyntaxNet Models for the CoNLL 2017 Shared Task},
author = {Chris Alberti and Daniel Andor and Ivan Bogatyy and Michael Collins and Dan Gillick and Lingpeng Kong and Terry Koo and Ji Ma and Mark Omernick and Slav Petrov and Chayut Thanapirom and Zora Tung and David Weiss},
journal= {arXiv preprint arXiv:1703.04929},
year = {2017}
}