Researchers have introduced PLOTTER, a novel framework designed to improve the coherence and logical consistency of narratives generated by large language models. Unlike existing methods that operate directly on text, PLOTTER plans narratives using structural graph representations, specifically event and character graphs. This approach allows for the evaluation, planning, and revision of narrative elements based on graph topology and logical constraints, optimizing the underlying structure before full text generation. Experiments show PLOTTER significantly outperforms current baselines in generating complex and coherent narratives. AI
Summary written by None from 1 source. How we write summaries →
IMPACT Enhances LLM capabilities in complex narrative generation by improving coherence and logical consistency.
RANK_REASON Academic paper introducing a new framework for narrative generation.
