Re3: Generating Longer Stories With Recursive Reprompting and Revision
This addresses the challenge of maintaining coherence and relevance in long-form story generation for applications in creative writing and content creation, representing an incremental improvement over existing methods.
The paper tackles the problem of automatically generating longer stories over 2,000 words by addressing challenges in plot coherence and relevance, achieving a 14% absolute increase in coherent plot and 20% increase in premise relevance compared to baseline methods.
We consider the problem of automatically generating longer stories of over two thousand words. Compared to prior work on shorter stories, long-range plot coherence and relevance are more central challenges here. We propose the Recursive Reprompting and Revision framework (Re3) to address these challenges by (a) prompting a general-purpose language model to construct a structured overarching plan, and (b) generating story passages by repeatedly injecting contextual information from both the plan and current story state into a language model prompt. We then revise by (c) reranking different continuations for plot coherence and premise relevance, and finally (d) editing the best continuation for factual consistency. Compared to similar-length stories generated directly from the same base model, human evaluators judged substantially more of Re3's stories as having a coherent overarching plot (by 14% absolute increase), and relevant to the given initial premise (by 20%).