RSTGen: Imbuing Fine-Grained Interpretable Control into Long-FormText Generators

TL;DR

This paper introduces RSTGen, a framework that leverages Rhetorical Structure Theory to enhance control over discourse, semantics, and topics in long-form text generation, improving coherence and interpretability.

Contribution

We propose RSTGen, a novel method that integrates Rhetorical Structure Theory into language models to enable fine-grained control over generated long-form text.

Findings

Model effectively controls discourse and semantic features.

Performs competitively on argument and story generation tasks.

Offers significantly more control over generated content.

Abstract

In this paper, we study the task of improving the cohesion and coherence of long-form text generated by language models. To this end, we propose RSTGen, a framework that utilises Rhetorical Structure Theory (RST), a classical language theory, to control the discourse structure, semantics and topics of generated text. Firstly, we demonstrate our model's ability to control structural discourse and semantic features of generated text in open generation evaluation. Then we experiment on the two challenging long-form text tasks of argument generation and story generation. Evaluation using automated metrics and a metric with high correlation to human evaluation, shows that our model performs competitively against existing models, while offering significantly more controls over generated text than alternative methods.