待验证50% 置信事实精确时间
Generative Agents研究论文标题为《Generative Agents: Interactive Simulacra of Human Behavior》，每个虚拟居民具备记忆流、反思机制和规划模块三个核心模块
1
来源数
50%
置信度
长期有效
时效性
2026/8/31
首次发现
来源
涉及实体
相关事实
待验证斯坦福2023年「虚拟小镇」（Generative Agents）实验以25个 GPT-4 实例模拟社会行为，展示了 LLM 多智能体的涌现协作能力68% 相似待验证微软的AutoGen框架和斯坦福的Generative Agents论文（Smallville实验，让25个Agent模拟人类社会行为）是多智能体系统方向的标志性工作68% 相似待验证Autonomous Agents possess an independent Perception-Reasoning-Action Loop originating from classic paradigms in cognitive science and robotics.68% 相似待验证Agent采用感知层、思考层、行动层三层架构协同运作67% 相似待验证Since 2023, projects like Stanford's Generative Agents, Microsoft's AutoGen, and Andrew Ng's Agentic Workflow have demonstrated that LLM-based agents can handle complex cognitive tasks including planning, execution, and reflection67% 相似
引用此条事实
Stable URI
https://kongchang.com/claim/832306API
curl https://kongchang.com/api/v1/knowledge/claims/832306MCP
get_claim(id=832306)