Re3: Generating Longer Stories With Recursive Reprompting and Revision

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 (Re 3 ) 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 Re 3 's stories as having a coherent overarching plot (by 14% absolute increase), and relevant to the given initial premise (by 20%). Autoregressive Context AI, Open Philanthropy, DARPA under the Se-maFor program (HR00112020054), the Machine Common Sense (MCS) program under Cooperative Agreement N66001-19-2-4032, and the NSF through a fellowship to the first author. The content does not necessarily reflect the position or the policy of the government, and no official endorsement should be inferred. on Empirical Methods in Natural Language Processing (EMNLP).