Unsupervised Learning of Discourse Structures using a Tree Autoencoder

TL;DR

This paper introduces an unsupervised, autoencoder-based method for generating discourse trees, aiming to improve the robustness and diversity of discourse structures across domains for better NLP task performance.

Contribution

It presents a novel unsupervised approach to discourse tree induction that is task-agnostic and capable of producing larger, more diverse discourse treebanks.

Findings

Effective inference of discourse trees across multiple domains

Improved quality and diversity of discourse structures

Potential enhancement for downstream NLP tasks

Abstract

Discourse information, as postulated by popular discourse theories, such as RST and PDTB, has been shown to improve an increasing number of downstream NLP tasks, showing positive effects and synergies of discourse with important real-world applications. While methods for incorporating discourse become more and more sophisticated, the growing need for robust and general discourse structures has not been sufficiently met by current discourse parsers, usually trained on small scale datasets in a strictly limited number of domains. This makes the prediction for arbitrary tasks noisy and unreliable. The overall resulting lack of high-quality, high-quantity discourse trees poses a severe limitation to further progress. In order the alleviate this shortcoming, we propose a new strategy to generate tree structures in a task-agnostic, unsupervised fashion by extending a latent tree induction framework with an auto-encoding objective. The proposed approach can be applied to any tree-structured objective, such as syntactic parsing, discourse parsing and others. However, due to the especially difficult annotation process to generate discourse trees, we initially develop a method to generate larger and more diverse discourse treebanks. In this paper we are inferring general tree structures of natural text in multiple domains, showing promising results on a diverse set of tasks.