MoPS: Modular Story Premise Synthesis for Open-Ended Automatic Story Generation

TL;DR

MoPS introduces a modular approach to automatically generate diverse, high-quality story premises by breaking down story elements, extracting key design paths, and synthesizing coherent premises with improved originality and completeness.

Contribution

We propose a novel modular framework for story premise synthesis that enhances diversity and quality over existing methods, enabling scalable automatic story generation.

Findings

Synthesized premises outperform large language models in diversity and originality.

Generated stories from MoPS premises show higher quality and coherence.

Our approach significantly reduces costs and increases scalability in story premise creation.

Abstract

A story premise succinctly defines a story's main idea, foundation, and trajectory. It serves as the initial trigger in automatic story generation. Existing sources of story premises are limited by a lack of diversity, uneven quality, and high costs that make them difficult to scale. In response, we introduce Modular Story Premise Synthesis (MoPS) which breaks down story premises into modules like background and persona for automated design and generation. MoPS consists of three phases: (1) Precollect a consistent set of candidates for each module to form a nested dictionary. (2) Extract a key path from the nested dictionary as the premise design. (3) Instruct an LLM to integrate the design into a coherent premise sentence. Thorough evaluations demonstrate that our synthesized premises excel in diversity, fascination, completeness, and originality compared to those induced from large language models and captured from public story datasets. Similarly, the extended novels and scripts generated from our premises also exhibit higher quality. In supplementary materials, we provide the MoPS code suite, along with 7.6k generated premises and 1k extended stories. Code: https://github.com/GAIR-NLP/MoPS.