Neural Combinatory Constituency Parsing

TL;DR

This paper introduces two fast neural models for constituency parsing that decompose the process into classification and vector composition, achieving high accuracy and efficiency on multiple languages.

Contribution

The paper presents novel binary and multi-branching neural models with theoretical and empirical efficiency improvements for constituency parsing.

Findings

Binary model achieves 92.54 F1 on Penn Treebank

Models with XLNet attain near state-of-the-art accuracy

Observed language-specific syntactic tendencies during training

Abstract

We propose two fast neural combinatory models for constituency parsing: binary and multi-branching. Our models decompose the bottom-up parsing process into 1) classification of tags, labels, and binary orientations or chunks and 2) vector composition based on the computed orientations or chunks. These models have theoretical sub-quadratic complexity and empirical linear complexity. The binary model achieves an F1 score of 92.54 on Penn Treebank, speeding at 1327.2 sents/sec. Both the models with XLNet provide near state-of-the-art accuracies for English. Syntactic branching tendency and headedness of a language are observed during the training and inference processes for Penn Treebank, Chinese Treebank, and Keyaki Treebank (Japanese).