共享世界，私有心智：面向长篇写作的结构化记忆作为世界构建

中文摘要 AI 辅助

撰写长篇小说的LLM智能体需要显式记忆不断演变的故事世界，以使新事件与既定事实保持一致。这种记忆必须区分异构的叙事信息，跨粒度整合故事发展，并恢复写作请求中隐含的依赖关系。我们提出NarraWorld，一个面向长篇写作的结构化记忆系统，将记忆构建视为世界创建。从共享的基于证据的图出发，NarraWorld衍生出四个相互连接的视图：世界事实、每个角色的信念、开放发展和假设分支（可能世界的延续）。具有原子闭合的分层聚合将事件整合为场景、情节线和情节，使每个更高级别的节点可追溯到其组成源片段。对于检索，计划重建从当前叙事情境和记忆预览中推断查询的依赖关系，然后在令牌预算内组装相关记录。在三个写作基准上，NarraWorld取得了最强的综合结果。其记忆还可迁移到情境化角色扮演，并在通用长期记忆基准上基本保持召回率，为智能体在多样化叙事任务中维持连贯故事世界铺平了道路。

英文摘要

LLM agents that write long-form fiction need an explicit memory of the evolving storyworld to keep new events consistent with established facts. Such memory must keep heterogeneous narrative information distinct, integrate story developments across granularities, and recover dependencies that a writing request leaves implicit. We present NarraWorld, a structured memory system for long-form writing that treats memory construction as world creation. From a shared evidence-grounded graph, NarraWorld derives four connected views: world facts, per-character beliefs, open developments, and hypothetical branches (possible-world continuations). Hierarchical aggregation with atomic closure consolidates events into scenes, plotlines, and plots, keeping each higher-level node traceable to its constituent source spans. For retrieval, planned reconstruction infers a query's dependencies from the current narrative situation and a preview of memory, then assembles the relevant records within a token budget. Across three writing benchmarks, NarraWorld achieves the strongest aggregate results. Its memory also transfers to situated role-playing and largely preserves recall on a general-purpose long-term memory benchmark, paving the way for agents that sustain coherent storyworlds across diverse narrative tasks.

发表机构

• Southeast University（东南大学）
• Beijing Zhongguancun Academy（北京中关村学院）
• Duke University（杜克大学）
• East China Normal University（华东师范大学）
• Renmin University of China（中国人民大学）
• Communication University of China（中国传媒大学）
• Jiangsu Second Normal University（江苏第二师范学院）
• ZhuiWen Technology Co., Ltd.（追闻科技有限公司）

机构由 AI 辅助整理，请以论文原文为准。