arXiv cs.CL· Siyang He, Qiqi Wang, Fan Liu, Yuncheng Huang, Pengfei Xu, Xi Chen, Ziwei He, Xipeng Qiu·· 12 小时前AI 评分31
DraMaze：面向多样化微短剧剧本生成的多智能体流水线与未来感知记忆机制
DraMaze: A Multi-Agent Pipeline with Future-Aware Memory for Diverse Microdrama Script Generation
AI 导读
DraMaze 是一个多智能体流水线，通过套路记忆引导的进化树搜索与变异探索多样化因果路径，并构建反向叙事依赖树按未来叙事目标组织与检索历史记忆。在最长 60 集、6 万词的剧本上，DraMaze 将因果模式重复率降低 32.1%，Logic 得分达 82.5，超过最佳基线 8.5 分，三个骨干 LLM 上均有一致提升。
正文 · 原文
Abstract:The rapid growth of microdramas has increased demand for high-quality script production. Long-form microdrama script generation with LLMs requires both diverse plot progression and long-range narrative coherence. Our observations reveal that candidate continuations increasingly concentrate on fewer narrative mechanisms as generation progresses, while conventional retrieval may reinforce trope reuse and overlook historical information essential to future plots. To address these challenges, we propose DraMaze, a multi-agent pipeline that explores diverse causal paths through trope memory-guided evolutionary tree search and mutation. It further constructs a Backward Narrative Dependency Tree to organize, retain, and retrieve historical memories based on future narrative goals. We also introduce MiDramaBench, a bilingual benchmark for evaluating script logic and creative quality across different script lengths, together with causal-pattern metrics for narrative repetition and diversity. Experiments show consistent improvements across three backbone LLMs. For 60-episode scripts of up to 60k words, DraMaze reduces causal-pattern repetition by 32.1% and achieves a Logic score of 82.5, outperforming the best-performing baseline by 8.5 points.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.32002 [cs.CL] |
| (or arXiv:2610.32002v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.32002 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Qiqi Wang [view email]
[v1]
Thu, 22 Oct 2026 23:54:48 UTC (3,468 KB)
来源：arXiv cs.CL · arxiv.org