Content Planning for Neural Story Generation with Aristotelian Rescoring

TL;DR

This paper introduces a content planning system for neural story generation that uses Aristotelian principles to improve global coherence and relevance, resulting in higher quality narratives.

Contribution

It presents a novel plot-generation model combined with Aristotelian-inspired rescoring models to enhance story structure and coherence in neural narrative generation.

Findings

Stories are more relevant to prompts with content planning.

Generated stories have higher overall quality.

Principled plot-structure improves coherence.

Abstract

Long-form narrative text generated from large language models manages a fluent impersonation of human writing, but only at the local sentence level, and lacks structure or global cohesion. We posit that many of the problems of story generation can be addressed via high-quality content planning, and present a system that focuses on how to learn good plot structures to guide story generation. We utilize a plot-generation language model along with an ensemble of rescoring models that each implement an aspect of good story-writing as detailed in Aristotle's Poetics. We find that stories written with our more principled plot-structure are both more relevant to a given prompt and higher quality than baselines that do not content plan, or that plan in an unprincipled way.