从事实到洞察：面向角色扮演智能体的角色驱动双记忆框架与数据集
From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
AI总结：
针对长期对话中角色扮演智能体因上下文窗口限制而丧失角色一致性的问题，提出角色记忆数据集RoleMemo和双记忆框架DualMem，通过将记忆解耦为事实认知和角色条件洞察，结合监督微调与强化学习，在4B参数模型上超越基于DeepSeek-V3.2的零样本角色无关框架。
摘要：
尽管角色扮演智能体在短期交互中表现出色，但长期对话会压垮上下文窗口，从而促使外部记忆框架的发展。当前系统通常依赖角色无关的摘要，记录事实而不进行角色特定的解释，导致生成通用回复，损害角色保真度。为弥补这一差距，我们引入了RoleMemo数据集，其中包含四个推理任务，这些任务要求通过角色解释事实片段以得出正确答案。在RoleMemo上的评估揭示了角色无关框架的关键局限性。因此，我们提出了DualMem，它将记忆解耦为两个流：事实认知和角色条件洞察。通过监督微调（SFT）和强化学习（RL）训练，我们的框架使用4B参数模型在持续角色保真度上优于由DeepSeek-V3.2驱动的零样本角色无关框架。我们的资源可在https://github.com/role2026/rolememo获取。
英文摘要：
While role-playing agents excel in short-term interactions, long-term conversations overwhelm context windows, motivating external memory frameworks. Current systems typically rely on persona-agnostic summarization, which records facts without persona-specific interpretation, yielding generic responses that compromise persona fidelity. To bridge this gap, we introduce RoleMemo, a dataset featuring four reasoning tasks where the factual fragments must be interpreted through the persona to reach the correct answer. Evaluation on RoleMemo exposes critical limitations of persona-agnostic frameworks. We thus propose DualMem, which decouples memory into two streams: factual cognition and persona-conditioned insight. Trained through Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL), our framework with a 4B-parameter model outperforms zero-shot persona-agnostic frameworks powered by DeepSeek-V3.2 for sustained persona fidelity. Our resources are available at https://github.com/role2026/rolememo.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
