[论文] Scaling Long-Form Story Generation via Narrative State Tracking

论文概要

研究领域: NLP
作者: Zhennan Wan, Jianfei Chen
发布时间: 2026-09-28
arXiv: 2609.35759

中文摘要

LLM 在创意写作方面已展现出强大能力，但将其扩展到全长小说仍然具有挑战性，因为保持叙事一致性变得越来越困难。现有故事生成方法通常专注于最多约一万字的故事，其扩展到全长小说的能力尚未充分探索。我们引入叙事状态追踪智能体（NstAgent），一个无需训练的智能体框架，允许 LLM 追踪结构化的叙事状态，包括角色、过去事件和未来需求。我们扩展了现有基准以跨长度比较叙事一致性，并将其与写作质量基准一起系统评估从 1 万字到 10 万字的故事。NstAgent 在故事变长时实现了更好的叙事一致性和写作质量，且两者均未随长度增加而明显退化，这表明它为将故事生成扩展到全长小说提供了有效途径。

原文摘要

LLMs have demonstrated strong capabilities in creative writing. However, scaling them to full-length novels remains challenging, as maintaining narrative consistency becomes increasingly difficult. Existing story-generation methods typically focus on stories of up to about ten thousand words, leaving their ability to scale to full-length novels underexplored. In this work, we introduce Narrative State Tracking Agent (NstAgent), a training-free agentic framework that allows LLMs to track a structured narrative state including characters, past events and future requirements. We extend an existing benchmark to compare narrative consistency across lengths, and use it together with a writing-quality benchmark to systematically evaluate stories ranging from 10K to 100K words. We show that NstAge...

自动采集于 2026-09-30

#论文 #arXiv #NLP #小凯