RSTGen: Imbuing Fine-Grained Interpretable Control into Long-FormText Generators
This addresses the challenge of fine-grained interpretable control in long-form text generation for applications like argument and story writing, though it is incremental as it builds on existing methods with a classical theory.
The paper tackles the problem of improving cohesion and coherence in long-form text generated by language models by proposing RSTGen, a framework that uses Rhetorical Structure Theory to control discourse structure, semantics, and topics. The model performs competitively against existing models in tasks like argument and story generation while offering significantly more control over generated text.
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.