让虚拟伙伴像真人一样随时间成长，情绪与行为相互影响。
Toward Natural and Companionable Virtual Agents via Cross-Temporal Emotional Modeling
- 用情绪状态闭环建模长期互动中的情感变化
- 21天真实使用测试中自然度、连贯性显著提升
- 适合想打造有情感深度的陪伴型智能体的研究者
近期基础模型的发展使对话代理具备了持续陪伴而非仅完成任务的能力。然而，大多数代理仍难以支持自然、长期的陪伴式交互，导致体验显得片段化且不真实。我们指出，当前代理忽略了对社会行为与内部情绪的跨时序建模：生成的行为很少影响代理的情绪状态，而情绪状态也极少塑造后续行为。本文提出跨时序情绪建模（CTEM）框架，将长期行为历史与即时情绪表达相连接。CTEM建立了一个闭环机制：过往经历更新动态情绪状态；该状态决定当前互动；用户反馈则持续修正记忆与情绪状态，实现反思与预期。我们以CTEM为基础构建了名为Auri的陪伴型代理，部署于即时通讯平台，并开展了为期21天的真实环境研究，结果表明，相较于基线，CTEM在感知自然性、连贯性和情绪和谐性方面均有显著提升。
原文摘要 · Abstract (English)
Recent advances in foundation models have enabled conversational agents that aim for sustained companionship rather than mere task completion. Yet most still remain unable to support natural, long-term companion-like interactions, resulting in experiences that feel episodic and inauthentic. We argue that current agents overlooked cross-temporal modeling of agents' social behaviors and internal emotions: generated behaviors rarely influence an agent's emotional state, and emotional states seldom shape subsequent behaviors. We present Cross-Temporal Emotion Modeling (CTEM), a framework that links long-term behavioral history to moment-to-moment emotional expression. CTEM establishes a closed loop where past experiences update an evolving emotional state; this state conditions immediate interactions; and user feedback continually revises both memory and emotional state, enabling reflection and anticipation. We instantiate CTEM as Auri, a companion agent on an instant-messaging platform, and report a 21-day in-the-wild study showing that CTEM shows improvements in perceived naturalness, coherence, and emotional harmony.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。