1、人工智能基础，聚焦联结主义、符号主义和行为主义的融合方法。主要研究问题：融合神经网络和符号方法的智能规划器；融合自然语言推理（LLMs）与形式化推理（LEAN、Isabelle、Coq、Z3等）的自动证明系统。

• Xingjin Wang, Howe Tissue, Lu Wang, Linjing Li, Daniel Zeng, "Entropy Scheduling in Reinforcement Learning for Large Language Models," in Findings of the 64^th Annual Meeting of the Association for Computational Linguistics (ACL), San Diego, California, USA, July 2 - 7, 2026.
• Xufeng Zhou, Linjing Li, and Daniel Zeng, "Token-level Advantage Policy Optimization from Negative Feedback in Multi-Turn Agents," in Proceedings of the 25th International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), Paphos, Cyprus，May 25 – 29, 2026.
• Songjun Tu, Jiahao Lin, Qichao Zhang, Xiangyu Tian, Linjing Li, Xiangyuan Lan, and Dongbin Zhao, "Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RL, " in Proceedings of the 39^th Annual Conference on Neural Information Processing Systems (NeurIPS), San Diego, USA, December 2 – 7, 2025. 
• Yifei Wang, Feng Xiong, Yong Wang, Linjing Li, Xiangxiang Chu, and Daniel Zeng, "POSITION BIAS MITIGATES POSITION BIAS: Mitigate Position Bias Through Inter-Position Knowledge Distillation," in Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP), Suzhou, China, November 5-9, 2025.
• Jiajing Zhang, Jiamei Jiang, Linjing Li, Chenyang Zhang, and Daniel Zeng, "A Closed-Loop Architecture with Knowledge-of-Results Feedback for Neural-Symbolic Planning," Knowledge-Based Systems (KBS), vol. 326, 114041, July 2025.
• Songjun Tu, Jiahao Lin, Xiangyu Tian, Qichao Zhang, Linjing Li, Yuqian Fu, Nan Xu, Wei He, Xiangyuan Lan, Dongmei Jiang, and Dongbin Zhao, "Enhancing LLM Reasoning with Iterative DPO: A Comprehensive Empirical Investigation," in Proceedings of the 2025 Conference on Language Modeling (COLM), Montreal, Canada, October 7-10, 2025.
• Xingjin Wang, Howe Tissue, Lu Wang, Linjing Li, and Daniel Zeng, "Learning Dynamics in Continual Pre-Training for Large Language Models," in Proceedings of the 2025 International Conference on Machine Learning (ICML), Vancouver, Canada, July 13 - 19, 2025. (Spotlight, Oral)
• Yifei Wang, Yu Sheng, Linjing Li,  and Daniel Zeng, "Uncertainty Unveiled: Can Exposure to More In-context Examples Mitigate Uncertainty for Large Language Models?," in Findings of the 63^rd Annual Meeting of the Association for Computational Linguistics (ACL), Vienna, Austria, July 27 – August 1st, 2025.
• Yu  Sheng, Wanting Wen, Linjing Li, and Daniel Zeng, "Evaluating  Generalization Capability of Language Models across Abductive, Deductive  and Inductive Logical Reasoning," in Proceedings of the 31st International Conference on Computational Linguistics (COLING), Abu Dhabi, UAE, January 19-24, 2025.
• Jiajing Zhang, Jiamei Jiang, Chenyang Zhang, Jiahui Shi, Chiyu Cai, Linjing Li, and Daniel Zeng, "A Symbolic Incubator for Training Planners via Recognition-based Feedback," in Proceedings of the 2025 IEEE International Conference on Intelligence and Security Informatics (ISI), HongKong, China, July 12-14, 2025.
• Yifei  Wang, Yuheng Chen, Wanting Wen, Yu Sheng, Linjing Li, and Daniel Zeng,  "Unveiling Factual Recall Behaviors of Large Language Models through  Knowledge Neurons," in Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP), Miami, Florida, USA, November 12-16, 2024.

2、AI for Science，基于规划、推理等AI技术，赋能各学科研究。

• 磐石·科学基础大模型，“人工智能+科学”的操作系统。
• Minghui Jia, Qichao Zhang, Ali Luo, Linjing Li, Shuo Ye, Hailing Lu, Wen Hou, and Dongbin Zhao, "Spec-o3: A Tool-Augmented Vision-Language Agent for Rare Celestial Object Candidate Vetting via Automated Spectral Inspection," in Proceedings of the 64^th Annual Meeting of the Association for Computational Linguistics (ACL), San Diego, California, USA, July 2 - 7, 2026.
• Zihan Lin, Zhi Fang, Jiayang Sun, Jie Cao, Huaibo Huang, Linjing Li, Junxian Duan, and Ran He, "Predictor–Corrector Flow Matching for Crystal Structure Generation," in Proceedings of the IEEE International Conference on Multimedia & Expo 2026 (ICME), Bangkok, Thailand, July 6-10 2026.
• Xinrui Wang, Chengbo Li, Boxuan Zhang, Jiahui Shi, Nian Ran, Linjing Li, Jianjun Liu, and Dajun Zeng, "S1-MatAgent: A planner driven multi-agent system for material discovery," arXiv:2509.14542, 2025.
• Zhaohui Yang, Chenghua He, Xiaowen Shi, Shihong Deng, Linjing Li, Qiyue Yin, and Daxin Jiang, "Beyond the First Error: Process Reward Models for Reflective Mathematical Reasoning," in Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP), Suzhou, China, November 5-9, 2025.
• Zhaohui Yang, Yuxiao Ye, Shilei Jiang, Shihong Deng, Chen Hu, Linjing Li, and Daxin Jiang, " Unearthing Gems from Stones: Policy Optimization with Negative Sample Augmentation for LLM Reasoning," in Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP), Suzhou, China, November 5-9, 2025.
• Yu  Sheng, Linjing Li,  and Daniel Zeng, "Learning Theorem Rationale for  Improving the  Mathematical Reasoning Capability of Large Language  Models," in Proceedings of the 39^th Annual AAAI Conference on Artificial Intelligence (AAAI), Philadelphia, Pennsylvania, USA, February 25 - March 4, 2025.
• Fan Yang, Jie Bai, Linjing Li, and Daniel Zeng, "Modeling Social Opinion Evolution with LLM Agents: Integrating Personality Traits with Embedded Opinion Dynamics," in Proceedings of the 2025 International Joint Conference on Neural Networks (IJCNN), Rome, Italy, June 30 - July 5, 2025.

3、博弈决策理论与方法，通过对博弈均衡及其特性的研究来为人工智能-博弈论交叉协同提供支撑。主要研究问题：均衡精炼及高效计算、均衡驱动的智能策略生成和优化、基于分布式人工智能（Agent）的博弈决策。

• Shiqi  Lei, Kanghoon Lee, Linjing Li, and Jinkyoo Park, "Learning Strategy  Representation for Imitation Learning in Multi-Agent Games," in Proceedings of the 39^th Annual AAAI Conference on Artificial Intelligence (AAAI), Philadelphia, Pennsylvania, USA, February 25 - March 4, 2025.
• Zhaohui Yang, Xiaoxuan Wang, and Linjing Li, "CPE: A New Paradigm for Policy Extraction in Offline Reinforcement Learning," in Proceedings of the 24^th International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), Detroit, Michigan, USA, May 19-23, 2025.
• Haorui  Li, Jiaqi Liang, Linjing Li, and Daniel Zeng, "Offline Meta  Reinforcement Learning with Weighted Policy Constraints and Proximal  Context Collection," in Proceedings of the 24^th International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), Detroit, Michigan, USA, May 19-23, 2025.
• Jiajing Zhang, Jiamei Jiang, Linjing Li, and Daniel Zeng, "A Novel Decision-Making Model for Playing Board Game Combining Planning and Opponent Behaviors," in Proceedings of the 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Hyderabad, India, April 6-11, 2025.
• Nuo Chen, Xuelong Liu, Linjing Li, and Wenji Mao, "Retrogressively Learning the Equilibrium Strategy of Pursuit-Evasion Games," in Proceedings of the 2025 IEEE International Conference on Intelligence and Security Informatics (ISI), HongKong, China, July 12-14, 2025.
• Nuo Chen, Linjing Li, and Wenji Mao, "Equilibrium Strategy of the Pursuit-Evasion Game in Three-Dimensional Space," IEEE/CAA Journal of Automatica Sinica, vol. 11, no. 2, pp. 446–458, February 2024.
• Shiqi  Lei, Kanghoon Lee, Linjing Li, Jinkyoo Park, and Jiachen Li, "ELA:  Exploited Level Augmentation for Offline Learning in Zero-Sum Games," in  Proceedings of the 23^th International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), Auckland, New Zealand, May 6-10, 2024.

要求报考学生在本科期间数学分析、线性代数、概率及统计成绩优秀，有较好的算法分析及设计基础，代码能力强(Python、Java)，英语听说读写无障碍。优先录取在以下方向有基础的学生。

• 数学：数理逻辑、泛函分析、抽象代数、拓扑、流形、最优化

• 物理：相对论、量子力学

• 计算机：理论计算机、形式化方法、编译原理、操作系统、分布式计算

• 控制论：最优控制、鲁棒控制、非线性控制

• 经济学：博弈论、宏/微观经济学、计量经济学、可计算一般均衡

2007-09--2011-07   中国科学院自动化研究所   工学博士
2005-09--2007-07   哈尔滨工业大学   工学硕士
2001-09--2005-07   哈尔滨工业大学   工学学士

   

2026-05~现在, 中国科学院自动化研究所, 研究员、博士研究生导师
2022-04~2026-04,中国科学院自动化研究所, 研究员，硕士研究生导师
2020-03~2022-03,中国科学院自动化研究所, 副研究员，硕士研究生导师
2014-11~2020-03,中国科学院自动化研究所, 副研究员
2012-07~2012-11,香港城市大学, 高级研究助理
2011-07~2014-10,中国科学院自动化研究所, 助理研究员

多智能体系统

课程名称：多智能体系统

主讲内容：基础博弈论

讲授时间：2021年春季、2022年春季、2023年春季

课程材料：GitHub