Long-form writing agents require flexible integration and interaction across information retrieval, reasoning, and composition. Current approaches rely on predetermined workflows and rigid thinking patterns to generate outlines before writing, resulting in constrained adaptability during writing. In this paper we propose a general agent framework that achieves human-like adaptive writing through recursive task decomposition and dynamic integration of three fundamental task types, i.e. retrieval, reasoning, and composition. Our methodology features: 1) a planning mechanism that interleaves recursive task decomposition and execution, eliminating artificial restrictions on writing workflow; and 2) integration of task types that facilitates heterogeneous task decomposition. Evaluations on both fiction writing and technical report generation show that our method consistently outperforms state-of-the-art approaches across all automatic evaluation metrics, which demonstrate the effectiveness and broad applicability of our proposed framework.

本研究解决了当前长篇写作代理在写作过程中灵活性不足的问题。我们提出了一种通用代理框架，通过递归任务分解和检索、推理、创作三种基本任务类型的动态整合，来实现类人自适应写作。实验证明，我们的方法在小说创作和技术报告生成中均优于当前最新的方法，展示了该框架的有效性和广泛适用性。

超越大纲：语言模型自适应长篇写作的异构递归规划