Retell, Reward, Repeat: Reinforcement Learning for Narrative Theory-Informed Story Retelling
arXiv:2601.17226v2 Announce Type: replace-cross Abstract: Counterfactual story retelling exposes LLM shortcomings in constrained narrative solution spaces where they can no longer rely on recalling memorised training data. Ground-truth-based post-training, such as SFT, fails to teach LLMs how to generate logical and rational narrative events. In this paper, we introduce Retell, Reward, Repeat (RRR), an RL-based pipeline synthesising Structuralist Narratology with scalar narrativity to teach storytelling structure. We extend the TimeTravel dataset with human-annotated stages of narrative equili
The increasing sophistication of LLMs and the recognition of their limitations in complex narrative generation are driving research into more robust storytelling methodologies.
This research addresses a fundamental limitation in generative AI – the ability to produce coherent, logical, and structurally sound narratives, which is critical for future autonomous AI applications.
The proposed RRR pipeline using RL and structuralist narratology offers a new paradigm for training LLMs beyond simple memorization, enabling more sophisticated and reliable story generation.
- · AI developers
- · Generative AI platforms
- · Entertainment industry
- · Education sector
- · Companies relying on naive LLM prompting
- · Current content farm models
Improved narrative capabilities in LLMs will enhance their utility in creative content generation and complex task execution.
More reliable storytelling AI could lead to new forms of interactive media, personalized learning experiences, and automated report generation.
The ability of machines to consistently generate logical narratives might accelerate the development of more trustworthy and human-aligned AI agents across various domains.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI