Large Discourse Treebanks from Scalable Distant Supervision

TL;DR

This paper introduces a framework for generating large-scale discourse treebanks using distant supervision from sentiment analysis, aiming to improve discourse parsing across diverse domains.

Contribution

It presents a novel method to create extensive discourse treebanks automatically, reducing reliance on limited human-annotated data.

Findings

Enables training of discourse parsers on larger datasets

Improves domain independence of discourse parsing models

Reduces need for manual annotation

Abstract

Discourse parsing is an essential upstream task in Natural Language Processing with strong implications for many real-world applications. Despite its widely recognized role, most recent discourse parsers (and consequently downstream tasks) still rely on small-scale human-annotated discourse treebanks, trying to infer general-purpose discourse structures from very limited data in a few narrow domains. To overcome this dire situation and allow discourse parsers to be trained on larger, more diverse and domain-independent datasets, we propose a framework to generate "silver-standard" discourse trees from distant supervision on the auxiliary task of sentiment analysis.