Reproduce “MoPS: Modular Story Premise Synthesis for Open-Ended Automatic Story Generation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yan Ma, Yu Qiao, Pengfei Liu
The paper introduces MoPS, a modular story premise synthesis framework that addresses the challenge of generating high-quality, diverse story premises for automatic story generation. MoPS decomposes premise creation into sequential modules and systematically combines them, resulting in premises that demonstrate superior diversity and quality compared to existing methods, and which subsequently improve the quality of generated stories.
Key insights
- Modular decomposition enhances LLM capabilities by structuring the generation process, mitigating issues like repetitiveness and inconsistent quality.
- The sequential dependency between modules ensures coherence by generating components based on previously selected elements.
- High-quality premise generation is foundational for effective evaluation and advancement of automatic story generation systems, enabling more accurate attribution of story quality issues.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.