Global Greedy Dependency Parsing

Abstract

Most syntactic dependency parsing models may fall into one of two categories: transition- and graph-based models. The former models enjoy high inference efficiency with linear time complexity, but they rely on the stacking or re-ranking of partially-built parse trees to build a complete parse tree and are stuck with slower training for the necessity of dynamic oracle training. The latter, graph-based models, may boast better performance but are unfortunately marred by polynomial time inference. In this paper, we propose a novel parsing order objective, resulting in a novel dependency parsing model capable of both global (in sentence scope) feature extraction as in graph models and linear time inference as in transitional models. The proposed global greedy parser only uses two arc-building actions, left and right arcs, for projective parsing. When equipped with two extra non-projective arc-building actions, the proposed parser may also smoothly support non-projective parsing. Using multiple benchmark treebanks, including the Penn Treebank (PTB), the CoNLL-X treebanks, and the Universal Dependency Treebanks, we evaluate our parser and demonstrate that the proposed novel parser achieves good performance with faster training and decoding.

SG-Net: Syntax-Guided Machine Reading Comprehension

“…Recently, dependency syntactic parsing have been further developed with neural network and attained new state-ofthe-art results (Zhang, Zhao, and Qin 2016;Li et al 2018;Ma et al 2018;Li, Zhao, and Parnow 2020). Benefiting from the highly accurate parser, neural network models could enjoy even higher accuracy gains by leveraging syntactic information rather than ignoring it (Chen et al 2017a;2017b;Duan et al 2019).…”

Span Model for Open Information Extraction on Accurate Corpus

“…In this work, we develop an Open IE system by adapting a modified span selection model which has been applied in Semantic Role Labeling (SRL) (Ouchi, Shindo, and Matsumoto 2018;He et al 2018b;Li et al 2019;Zhao, Zhang, and Kit 2013;Zhao, Chen, and Kit 2009;Li et al 2018b;Cai et al 2018), coreference resolution (Shou and Zhao 2012;Zhang, Wu, and Zhao 2012) and syntactic parsing (Zhang, Zhao, and Qin 2016;Ma and Zhao 2012;Li, Zhao, and Parnow 2020;Zhou and Zhao 2019;Li et al 2018a). The advantage of a span model is that span level features can be sufficiently exploited which is hard to perform in token based sequence labeling models.…”

Graphspeech: Syntax-Aware Graph Attention Network for Neural Speech Synthesis

“…Relation Encoder converts syntax tree of input text into a syntax graph that describes a global relationship among the involved input tokens. Syntactic dependency parse tree [26,27] is one of the traditional ways to describe linguistic dependency relation among words. In a tree structure, only words that are directed related in a sentence are connected.…”