发表机构
Hangzhou International Innovation Institute, Beihang University; School of Computer Science, Fudan University（北京航空航天大学杭州国际创新研究院; 复旦大学计算机科学技术学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
DREAM是一种受ABC认知模型启发的结构化记忆框架，通过EMG提升角色扮演智能体的时间与因果连贯性，在CoSER、LIFECHOICE和TCM上取得最优性能。
AI 中文摘要
角色扮演智能体（RPAs）已成为大语言模型的关键应用，可实现沉浸式、高保真的角色模拟。准确演绎既定角色不仅需要风格模仿，还需具备时间一致、因果关联的行为推理能力。然而，现有角色扮演智能体主要依赖静态角色描述与非结构化记忆，限制了其维持长期叙事与人格连贯性的能力。我们提出DREAM，一种受激活事件-信念-结果（ABC）认知模型启发的角色扮演智能体结构化记忆框架。DREAM将非结构化文学文本转化为事件感知记忆图（EMG），该图将角色经验组织为时间有序、因果关联的事件图。此表示方法可构建动态的双粒度角色档案，同时捕捉稳定人格特质与事件驱动的行为演变。我们进一步提出时间因果记忆（TCM）基准，用于评估时间一致性与长程因果叙事连贯性。DREAM在CoSER、LIFECHOICE和TCM上均达到了当前最优性能，优于多个强基线模型。我们的方法证明了结构化记忆在提升角色扮演智能体可解释性与连贯性方面的有效性。
英文摘要
Role-playing agents (RPAs) have emerged as a key application of large language models, enabling immersive and high-fidelity character simulation. Accurate role-playing of established characters requires not only stylistic imitation but also temporally consistent and causally grounded behavioral reasoning. However, existing RPAs primarily rely on static character descriptions and unstructured memory, limiting their ability to maintain long-term narrative and personality coherence. We introduce DREAM, a structured memory framework for role-playing agents inspired by the Activating Event-Belief-Consequence (ABC) cognitive model. DREAM transforms unstructured literary text into an Event-aware Memory Graph (EMG) that organizes character experiences into temporally ordered and causally linked event graph. This representation enables the construction of dynamic, dual-granularity character profiles that capture both stable personality traits and event-driven behavioral evolution. We further propose the Temporal Causal Memory (TCM) benchmark to evaluate temporal consistency and long-range causal narrative coherence. DREAM achieves state-of-the-art performance across CoSER, LIFECHOICE, and TCM, outperforming multiple strong baselines. Our approach demonstrates the effectiveness of structured memory in enhancing the interpretability and consistency of role-playing agents.
CommentsAccepted at KDD 2026. Camera-ready version to appear. 16 pages, 5 figures
