大会议程 | CCF多智能体系统会议（CCMAS 2026）

Two-Day ProgramTwo-Day Program

详细日程Detailed Schedule

Day 1Day 1

2026 年 6 月 27 日（周六）June 27, 2026 (Saturday)

主会场（08:30-12:20）Main Hall (08:30-12:20)

08:30-08:40开幕式Opening Ceremony
08:40-09:20If Multi-Agent Foundation Models is the Answer, What is the Question?If Multi-Agent Foundation Models is the Answer, What is the Question?Stefano V. Albrecht 南洋理工大学Stefano V. Albrecht Nanyang Technological University
09:20-10:00机器人基础模型的进展、机遇与挑战Progress, Opportunities and Challenges of Robot Foundation Models李航字节跳动Hang Li ByteDance
10:00-10:20茶歇交流Tea Break
10:20-11:00语言智能原生的人工智能Language Intelligence-Native Artificial Intelligence张民哈尔滨工业大学（深圳）Min Zhang Harbin Institute of Technology, Shenzhen
11:00-11:40通用世界模型：从视频生成到物理行动General World Models: From Video Generation to Physical Action朱军清华大学Jun Zhu Tsinghua University
11:40-12:20
圆桌讨论：大模型时代下多智能体研究：范式迭代、学术破局与研究者新坐标Panel Discussion: Multi-Agent Research in the Era of Large Models: Paradigm Iteration, Academic Breakthrough, and Researchers' New Coordinates
主持人： Moderator: 张伟楠（上海交通大学） Weinan Zhang (Shanghai Jiao Tong University)
嘉宾： Guests:
- 安波（南洋理工大学） Bo An (Nanyang Technological University)
- Stefano V. Albrecht（南洋理工大学） Stefano V. Albrecht (Nanyang Technological University)
- 李航（字节跳动） Hang Li (ByteDance)
- 张民（哈尔滨工业大学（深圳）） Min Zhang (Harbin Institute of Technology, Shenzhen)
- 朱军（清华大学） Jun Zhu (Tsinghua University)
- 胡舒悦（上海人工智能实验室） Shuyue Hu (Shanghai Artificial Intelligence Laboratory)

分论坛并行时段（14:00-17:20）Parallel Forum Window (14:00-17:20)

分论坛 1：大模型智能体Forum 1: LLM Agents

主持人：郑岩天津大学Organizer: Yan Zheng Tianjin University

14:00-14:30大模型智能体记忆系统的错误追踪与自主演化Error Tracking and Autonomous Evolution of LLM Agent Memory Systems张宁豫浙江大学Ningyu Zhang Zhejiang University
14:30-15:00SciMaster，你的 AI 科学家朋友SciMaster, your AI scientist friend陈思衡上海交通大学Siheng Chen Shanghai Jiao Tong University
15:00-15:30多模态自主进化智能体Multimodal Autonomous Evolutionary Agents张志忠华东师范大学Zhizhong Zhang East China Normal University
15:30-15:50茶歇交流Tea Break
15:50-16:20Darwin-Agent：自主演进的Agent系统Darwin-Agent: An Autonomously Evolving Agent System邵坤小米Kun Shao Xiaomi
16:20-16:50大模型对齐中的奖励学习：稳定性与可识别性Reward Learning in LLM Alignment: Stability and Identifiability陈阳上海人工智能实验室Yang Chen Shanghai AI Laboratory
16:50-17:20大模型智能体的安全风险与可信治理Security Risks and Trustworthy Governance of LLM Agents刘艺格里菲斯大学Yi Liu Griffith University

分论坛 2：多智能体强化学习Forum 2: Multi-Agent Reinforcement Learning

主持人：温睦宁上海交通大学Organizer: Muning Wen Shanghai Jiao Tong University

14:00-14:30大语言环境下的强化学习Reinforcement Learning in Large Language Environments温颖上海交通大学Ying Wen, Shanghai Jiao Tong University
14:30-15:00实证博弈论分析在多具身对抗中的应用Application of Empirical Game Theory Analysis in Multi-Embodied Confrontation周铭上海人工智能实验室Ming Zhou, Shanghai AI Laboratory
15:00-15:30从外部反馈到内部认知：双重机制提升LLM智能体决策质量与效率From External Feedback to Internal Cognition: Dual Mechanisms to Improve Decision Quality and Efficiency of LLM Agents叶方华腾讯Fanghua Ye, Tencent
15:30-15:50茶歇交流Tea Break
15:50-16:20Causal World Understanding in Open-Ended Worlds: From Causal Representation Learning to Self-Improving AgentsCausal World Understanding in Open-Ended Worlds: From Causal Representation Learning to Self-Improving Agents杨梦月布里斯托大学Mengyue Yang, University of Bristol
16:20-16:50面向开放世界的具身多智能体自适应协作Embodied Multi-Agent Adaptive Collaboration for Open Worlds李阳上海交通大学Yang Li, Shanghai Jiao Tong University
16:50-17:20执委工作会议Executive Committee Working Meeting

分论坛 3：博弈论与机制设计Forum 3: Game Theory and Mechanism Design

主持人：孔芳南方科技大学李斌南京理工大学Organizers: Fang Kong SUSTech Bin Li Nanjing University of Science and Technology

14:00-14:30AI驱动的流量变现：从智能决策到平台机制设计AI-Driven Traffic Monetization: From Intelligent Decision-Making to Platform Mechanism Design祁琦中国人民大学Qi Qi, Renmin University of China
14:30-15:00多智能体合作中的能力评估探索Exploration of Capability Evaluation in Multi-Agent Cooperation赵登吉上海科技大学Dengji Zhao, ShanghaiTech University
15:00-15:30Optimal Analysis for Bandit Learning in Matching MarketsOptimal Analysis for Bandit Learning in Matching Markets李帅上海交通大学Shuai Li, Shanghai Jiao Tong University
15:30-15:50茶歇交流Tea Break
15:50-16:20Randomized Strategyproof Mechanisms with Best of Both Worlds Fairness and EfficiencyRandomized Strategyproof Mechanisms with Best of Both Worlds Fairness and Efficiency孙安康山东大学Ankang Sun, Shandong University
16:20-16:50博弈学习动力学现象与分析Game Learning Dynamics: Phenomena and Analysis穆义芬中国科学院数学与系统科学研究院Yifen Mu, Academy of Mathematics and Systems Science, Chinese Academy of Sciences
16:50-17:20金融网络博弈中的均衡问题研究Research on Equilibrium Problems in Financial Network Games周昊南昌大学Hao Zhou, Nanchang University

晚间安排（18:30-20:00）Evening Session (18:30-20:00)

18:30-20:00晚宴交流（赞助商讲话）Banquet and Networking (with Sponsor Speech)

Day 2Day 2

2026 年 6 月 28 日（周日）June 28, 2026 (Sunday)

分论坛并行时段（09:00-12:20）Parallel Forum Window (09:00-12:20)

分论坛 4：具身智能Forum 4: Embodied Intelligence

主持人：霍静南京大学Organizer: Jing Huo Nanjing University

09:00-09:30世界模型驱动的具身智能基础模型World Model-driven Foundation Models for Embodied Intelligence朱正极佳科技Zheng Zhu Geek+ Technology
09:30-10:00空间智能与具身世界模型Spatial Intelligence and Embodied World Models高宸清华大学Chen Gao Tsinghua University
10:00-10:20茶歇交流Tea Break
10:20-10:50面向具身大模型的强化学习后训练Reinforcement Learning Post-training for Embodied Large Models李浩然中国科学院自动化研究所Haoran Li, Institute of Automation, Chinese Academy of Sciences
10:50-11:20具身模型的真机评测与数据配方Real-robot Evaluation and Data Recipes for Embodied Models李永露上海交通大学Yonglu Li Shanghai Jiao Tong University
11:20-11:50星际争霸II中的具身进化：基于大模型推理的决策范式研究Embodied Evolution in StarCraft II: Decision-making Paradigms Based on Large-model Reasoning杨以钦中国科学院自动化研究所Yiqin Yang, Institute of Automation, Chinese Academy of Sciences
11:50-12:20ORCA：世界基座模型——像人一样学习世界ORCA: World Foundation Model — Learning the World Like a Human王鹏伟北京智源Pengwei Wang Beijing Academy of Artificial Intelligence

分论坛 5：从模型训练到科学发现Forum 5: From Model Training to Scientific Discovery

主持人：胡舒悦浦江实验室Organizer: Shuyue Hu Pujiang Laboratory

09:00-09:30面向长周期自主科学研究的智能体框架An Agent Framework for Long-horizon Autonomous Scientific Research冯世阳上海人工智能实验室Shiyang Feng Shanghai AI Laboratory
09:30-10:00从游戏AI到Agentic LLM：我们到底需要什么样的agent？From Game AI to Agentic LLM: What Kind of Agent Do We Really Need?张晨上海人工智能实验室Chen Zhang Shanghai AI Laboratory
10:00-10:20茶歇交流Tea Break
10:20-10:50智能体强化学习中的信用分配机制Credit Assignment Mechanisms in Agent Reinforcement Learning赵梦辰华南理工大学Mengchen Zhao South China University of Technology
10:50-11:20Computer Science of Language ModelsComputer Science of Language Models王新润新加坡管理大学Xinrun Wang Singapore Management University
11:20-11:50基于多智能体强化学习的博弈占优策略探索Exploring Dominant Strategies in Games via Multi-Agent Reinforcement Learning苏奇上海交通大学Qi Su Shanghai Jiao Tong University
11:50-12:20多智能体强化学习神经元优化方法研究Research on Neuronal Optimization Methods in Multi-Agent Reinforcement Learning沈思淇厦门大学Siqi Shen Xiamen University

分论坛 6：多智能体与社会科学交叉研究Forum 6: Multi-Agent and Social Science Interdisciplinary Research

主持人：华东师范大学钱鸿王祥丰同济大学李文浩Organizers: Hong Qian, Xiangfeng Wang ECNU Wenhao Li Tongji University

09:00-09:30以人为中心的人智协同决策Human-centered Human-AI Collaborative Decision-making卢暾复旦大学Tun Lu Fudan University
09:30-10:00构建基于大模型智能体的社会科学实验场Building a Social Science Experiment Field Based on LLM Agents陈旭中国人民大学Xu Chen Renmin University of China
10:00-10:20茶歇交流Tea Break
10:20-10:50基于LLM Agents的学习者建模研究Learner Modeling Based on LLM Agents陈静远浙江大学Jingyuan Chen Zhejiang University
10:50-11:20AI智能体安全：近端漏洞发现与远端风险推演AI Agent Safety: Proximal Vulnerability Discovery and Long-range Risk Simulation邵婧上海人工智能实验室Jing Shao Shanghai AI Laboratory
11:20-11:50教育智能体的构建与应用探索Construction and Application Exploration of Educational Agents沈双宏合肥综合性国家科学中心人工智能研究院Shuanghong Shen, Institute of Artificial Intelligence, Hefei Comprehensive National Science Center
11:50-12:20面向具身协同的多智能体强化学习方法Multi-Agent Reinforcement Learning Methods for Embodied Collaboration袁雷南京大学Lei Yuan Nanjing University