Joint Chinese Word Segmentation and Span-based Constituency Parsing

TL;DR

This paper introduces a joint Chinese word segmentation and span-based constituency parsing method that improves parsing accuracy by integrating segmentation directly into the parsing process, reducing errors caused by separate segmentation steps.

Contribution

It proposes a novel approach that adds extra labels to characters for joint segmentation and parsing, enhancing performance over existing models.

Findings

Outperforms recent joint segmentation and parsing models on CTB 5.1

Reduces errors caused by separate segmentation and parsing steps

Demonstrates effectiveness of label augmentation in joint modeling

Abstract

In constituency parsing, span-based decoding is an important direction. However, for Chinese sentences, because of their linguistic characteristics, it is necessary to utilize other models to perform word segmentation first, which introduces a series of uncertainties and generally leads to errors in the computation of the constituency tree afterward. This work proposes a method for joint Chinese word segmentation and Span-based Constituency Parsing by adding extra labels to individual Chinese characters on the parse trees. Through experiments, the proposed algorithm outperforms the recent models for joint segmentation and constituency parsing on CTB 5.1.