Zero-shot Chinese Discourse Dependency Parsing via Cross-lingual Mapping

TL;DR

This paper introduces a zero-shot Chinese discourse dependency parsing method that leverages English labeled data and cross-lingual mapping, enabling Chinese parsing without extensive labeled Chinese datasets.

Contribution

The paper proposes a novel cross-lingual mapping approach for zero-shot discourse parsing in Chinese using English data and translation-based parsing results.

Findings

Effective Chinese discourse trees generated without Chinese labeled data

Cross-lingual mapping from English to Chinese is feasible for discourse parsing

Method simplifies discourse parsing in low-resource languages

Abstract

Due to the absence of labeled data, discourse parsing still remains challenging in some languages. In this paper, we present a simple and efficient method to conduct zero-shot Chinese text-level dependency parsing by leveraging English discourse labeled data and parsing techniques. We first construct the Chinese-English mapping from the level of sentence and elementary discourse unit (EDU), and then exploit the parsing results of the corresponding English translations to obtain the discourse trees for the Chinese text. This method can automatically conduct Chinese discourse parsing, with no need of a large scale of Chinese labeled data.