Joint Syntacto-Discourse Parsing and the Syntacto-Discourse Treebank

TL;DR

This paper introduces the first end-to-end joint syntacto-discourse parser and a combined treebank, achieving state-of-the-art accuracy without preprocessing by integrating syntax and discourse parsing.

Contribution

It presents a novel joint parsing model and a combined treebank that unify syntactic and discourse analysis in an end-to-end framework.

Findings

Achieves state-of-the-art end-to-end discourse parsing accuracy

Requires no preprocessing such as segmentation or feature extraction

First to integrate Penn Treebank with RST Treebank in a joint parser

Abstract

Discourse parsing has long been treated as a stand-alone problem independent from constituency or dependency parsing. Most attempts at this problem are pipelined rather than end-to-end, sophisticated, and not self-contained: they assume gold-standard text segmentations (Elementary Discourse Units), and use external parsers for syntactic features. In this paper we propose the first end-to-end discourse parser that jointly parses in both syntax and discourse levels, as well as the first syntacto-discourse treebank by integrating the Penn Treebank with the RST Treebank. Built upon our recent span-based constituency parser, this joint syntacto-discourse parser requires no preprocessing whatsoever (such as segmentation or feature extraction), achieves the state-of-the-art end-to-end discourse parsing accuracy.