万海
教授
联系邮箱: wanhai@mail.sysu.edu.cn
联系地址: A220, 计算机学院, 中山大学 (东校区)
研究领域:
人工智能基础理论、形式化方法.
具体包括:知识的表示与推理(KR)、可信大模型、可满足性问题 (SAT)、组合优化问题 (COP)等.
工作经历:
- 2022.4-至今 中山大学计算机学院(软件学院),教授,博导.
- 2016.2-2022.4 中山大学计算机学院(软件学院),副教授,博导.
科研项目:
- 2022年国家自然科学基金;双向可解释的线性时态逻辑和深度神经网络融合推理理论研究与应用;项目批准号:62276284;在研;万海(主持人).
- 2019年国家自然科学基金;基于概率稳定模型语义溯因推理机制的可解释视觉问答研究与实现;项目批准号:61976232;在研;万海(主持人).
- 2015年国家自然科学基金;稳定模型语义下不一致本体的存在规则语言修复理论研究与实现;项目批准号:61573386;已结题;万海(主持人).
- 国家重点研发计划科研项目子课题;语义信息论中关键问题的数学基础与方法;项目批准号:2021YFA1000504;在研;万海(主持人).
- 国家重点研发计划科研项目子课题;多源异构数据的案件特征与业务态势分析;项目批准号:2018YFC0830600;已结题;万海(主持人).
- 2017年广东省前沿与关键技术创新项目;虚拟现实与增强现实场景下基于多模态人机自然交互的3D建模技术与产业化应用;项目批准号:2017B010110011;已结题;万海(主持人).
- 2016年广东省前沿与关键技术创新项目;面向教育的分布式云系统-云簇关键技术研究;项目批准号:2016B030305007;已结题;万海(主持人).
- 2022年广东省自然科学基金;基于线性时态逻辑的深度神经网络增强模型理论研究及在可信人工智能的应用;项目批准号:2022A1515011355;在研;万海(主持人).
- 2016年广东省自然科学基金;面向公共知识的多智能体认知规划关键技术研究与实现;项目批准号:2016A030313292;已结题;万海(主持人).
- 2012年广东省自然科学基金;若干非单调逻辑可译问题研究与求解器实现;项目批准号:S2012010009836;已结题;万海(主持人).
- 2011年度博士点基金自然科学类课题;稳定模型下一阶逻辑理论求解及其在基于模型程序诊断的应用研究;项目批准号:20110171120041;已结题;万海(主持人).
- 2018年广州市科技计划项目;稳定模型语义下存在规则语言的本体查询回答理论研究与实现;项目批准号:201804010435;已结题;万海(主持人).
- 2013年广州市科技计划项目;若干非单调逻辑可译性研究及其在面向SOA服务组合诊断;项目批准号:2013J41000582013;已结题;万海(主持人).
- 2016高校基本科研业务费中山大学青年教师培育项目;有穷结构上若干一阶非单调逻辑计算方法的研究及异构海量数据语义查询求解器实现;已结题;万海(主持人).
- 2016年科技服务项目;广西壮族自治区推进机器人产业发展基本思路;已结题;万海(主持人).
主要学术兼职:
- 中国计算机学会理论计算机专委会执行委员.
- 中国计算机学会形式化方法专委会执行委员.
- 中国计算机学会人工智能与模式识别专委会执行委员会正式委员.
- 中国中文信息学会知识图谱与知识计算专委会执行委员.
- 广东省本科高校计算机类专业教学指导委员会秘书长.
- 中国计算机学会NOI广东省特派员.
教授课程:
- 研究生课程:计算机程序理论与模型、知识的表示与推理.
- 本科生课程:人工智能、编译原理、操作系统、程序设计、自主移动机器人导论、自然语言处理与知识图谱.
代表性论著:
中国计算机协会CCF推荐A类和B类论文
- HaiWan(万海), Biqing Fang, Yongmei Liu. A General Multi-agent Epistemic Planner Based on Higher-order Belief Change [J]. Artificial Intelligence Journal, 2021, 301: 103562.
- Yanan Liu, Hai Wan(万海,通讯作者), Jianfeng Du, Yao Wang, Kunxun Qi, Weilin Luo. Learning to mine all minimal evidences for unverified claims[J]. Information Sciences, 2025, 720: 122505.
- Weilin Luo, Polong Chen, Hai Wan(万海,通讯作者), Shaowei Cai, Zhanhao Xiao. Goal-conflict identification based on local search and fast boundary-condition verification based on incremental satisfiability filter[J]. Journal of Systems and Software, 2024, 213: 112036.
- Weilin Luo, Pingjia Liang, Jianfeng Du, Hai Wan(万海,通讯作者), Bo Peng, Delong Zhang. Bridging LTLf Inference to GNN Inference for Learning LTLf Formulae[C]//In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI 2022), 2022, 36(9): 9849-9857.
- Hai Wan(万海), Manrong Zhang, Jianfeng Du, Ziling Huang, Yufei Yang, Jeff Z. Pan. FL-MSRE: A Few-Shot Learning based Approach to Multimodal Social Relation Extraction [C]//In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI 2021), 2021: 13916-13923.
- Weilin Luo, Hai Wan(万海,通讯作者), Rongzhen Ye, Shaowei Cai, Biqing Fang, Delong Zhang: Improving Local Search Algorithms via Probabilistic Configuration Checking[C]//In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI 2022), 2022: 10283-10290.
- Rongzhen Ye, Tianqu Zhuang, Hai Wan(万海,通讯作者), Jianfeng Du, Weilin Luo, Pingjia Liang. A Noise-Tolerant Differentiable Learning Approach for Single Occurrence Regular Expression with Interleaving[C]//In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI 2023), 2023, 37(4): 4809-4817.
- Hai Wan(万海), Pingjia Liang, Jianfeng Du, Weilin Luo, Rongzhen Ye, Bo Peng. End-to-end learning of ltlf formulae by faithful ltlf encoding[C]//In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI 2024), 2024, 38(8): 9071-9079.
- Hai Wan(万海), Haicheng Chen, Jianfeng Du, Weilin Luo, Rongzhen Ye. A DQN-based Approach to Finding Precise Evidences for Fact Verification[C]//In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL/IJCNLP 2021), 2021, 1: 1030-1039.
- Hai Wan(万海), Jinxia Lin, Jianfeng Du, Dawei shen, Rongman Zhang. Enhancing Metaphor Detection by Gloss-based Interpretations[C]//In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL/IJCNLP 2021), 2021: 1971-1981.
- Kunxun Qi, Hai Wan(万海,通讯作者), Jianfeng Du, Haolan Chen: Enhancing Cross-lingual Natural Language Inference by Prompt-learning from Cross-lingual Templates[C]//In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL/IJCNLP 2022), 2022: 1910-1923
- Kunxun Qi, Jianfeng Du, Hai Wan(万海,通讯作者). End-to-end Learning of Logical Rules for Enhancing Document-level Relation Extraction[C]//In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL/IJCNLP 2024), 2024, 1: 7247-7263.
- Weilin Luo, Hai Wan(万海,通讯作者), Delong Zhang, Hengdi Su. Checking ltl satisfiability via end-to-end learning[C]//In Proceedings of the IEEE/ACM International Conference on Automated Software Engineering (ASE 2022), 2022, 21: 1-13.
- Bo Peng, Pingjia Liang, Tingchen Han, Weilin Luo, Jianfeng Du, Hai Wan(万海,通讯作者), Rongzhen Ye, Yuhang Zheng. Purltl: Mining ltl specification from imperfect traces in testing[C]//In Proceedings of the IEEE/ACM International Conference on Automated Software Engineering (ASE 2023), 2023: 1766-1770.
- Weilin Luo, Yuhang Zheng, Rongzhen Ye, Hai Wan(万海,通讯作者), Jianfeng Du, Polong Chen. SAT-Verifiable LTL Satisfiability Checking via Graph Representation Learning[C] //In Proceedings of the IEEE/ACM International Conference on Automated Software Engineering (ASE 2023), 2023: 1761-1765.
- Kunxun Qi, Jianfeng Du, Hai Wan(万海,通讯作者). Bi-directional Learning of Logical Rules with Type Constraints for Knowledge Graph Completion[C]//In Proceedings of the ACM International Conference on Information and Knowledge Management (CIKM 2024), 2024: 1899-1908.
- Junming Qiu, Rongzhen Ye, Weilin Luo, Kunxun Qi, Hai Wan(万海,通讯作者), Yue Yu. OBDD-NET: End-to-End Learning of Ordered Binary Decision Diagrams[C]//In Proceedings of the ACM International Conference on Information and Knowledge Management (CIKM 2025), 2025: 2430-2439.
- Weilin Luo, Hai Wan(万海,通讯作者), Hongzhen Zhong, Ou Wei, Biqing Fang, Xiaotong Song. An efficient two-phase method for prime compilation of non-clausal boolean formulae[C]//In Proceedings of the IEEE/ACM International Conference On Computer Aided Design (ICCAD 2021), 2021: 1-9.
- Weilin Luo, Hai Wan(万海,通讯作者), Xiaotong Song, Binhao Yang, Hongzhen Zhong, Yin Chen. How to Identify Boundary Conditions with Contrasty Metric?[C] Proceedings of the IEEE/ACM International Conference on Software Engineering (ICSE 2021), 2021: 1473-1484.
- Weilin Luo, Weiyuan Fang, Junming Qiu, Hai Wan(万海,通讯作者), Yanan Liu, Rongzhen Ye. ITG: Trace Generation via Iterative Interaction between LLM QueryandTraceChecking[C]//In Proceedings of the ACM/IEEE International Conference on Software Engineering (ICSE 2024), 2024: 11-15.
- Weilin Luo, Hai Wan(万海,通讯作者), Jianfeng Du, Xiaoda Li, Yuze Fu, Rongzhen Ye, Delong Zhang. Teaching LTLf Satisfiability Checking to Neural Networks[C]//In Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI 2022), 2022: 3292-3298.
- Weilin Luo, Pingjia Liang, Junming Qiu, Polong Chen, Hai Wan(万海,通讯作者), Jianfeng Du, Weiyuan Fang. Learning to Check LTL Satisfiability and to Generate Traces via Differentiable Trace Checking[C]//In Proceedings of the ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA 2024), 2024: 996-1008.
- Weilin Luo, Tingchen Han, Junming Qiu, Hai Wan(万海,通讯作者), Jianfeng Du, Bo Peng, Guohui Xiao, Yanan Liu. NADA: Neural Acceptance-Driven Approximate Specification Mining[C]//In Proceedings of the ACM on Software Engineering (ISSTA 2025), 2025, 2: 1795-1817.
- Kunxun Qi, Jianfeng Du, Hai Wan(万海,通讯作者). Learning from Both Structural and Textual Knowledge for Inductive Knowledge Graph Completion[C]//In Proceedings of Advances in Neural Information Processing Systems (NeurIPS 2023), 2023, 36: 26546-26562.
已授权知识产权
- 万海,罗炜麟,陈泼龙,曾娟.一种用于嵌入式系统存储功能测试的数据增广方法及系统: 中国, CN118885400B [P]. 2024-12-13.
- 万海,罗炜麟,邓浩天,梁梓豪,曾娟. 一种基于有限线性时态逻辑无监督可满足性检测的软件设计方法及系统: 中国, CN118796163B [P]. 2024-12-13.
- 万海,刘亚男,张漫榕,黄佳莉,曾娟.一种基于时态知识图谱推理的社交关系演变预测方法: 中国, CN112905738B [P]. 2022-04-22.
- 万海,钟洪桢,罗炜麟,范科峰,赵杭. 一种形式化需求规约自动化修复方法: 中国, CN112667215B [P]. 2022-02-25.
- 万海,欧佳玲,曾娟,王宝亿. 面向移动端虚拟现实与增强现实的场景图谱生成方法: 中国, CN111144492B [p]. 2023-03-28.
- 万海,罗永豪. 一种基于图像场景图谱对齐的图像查询回答方法: 中国, CN109800317B [P]. 2022-03-11.
- 万海,肖逸凡,曾娟.一种基于迭代精细化的图像场景图谱生成方法: 中国, CN109783666B [P]. 2023-05-23.
- 万海,夏勇涛,曾娟.一种基于图二阶相似性的知识图谱实体语义空间嵌入方法: 中国, CN109829057B [P]. 2023-02-21.
- 万海,刘欣怡.基于源代码多标签图神经网络的程序代码抄袭类型检测方法与系统: 中国, CN108446540B [P]. 2022-02-25.
- 曾娟,刘亚男,方维远,罗炜麟,万海. 基于大语言模型的控制系统自动化修复方法及系统: 中国, CN120065871B [P]. 2026-03-20.
- 曾娟,彭勃,罗炜麟,李耀华,万海. 一种基于神经符号融合算法的数值推理问题弱监督求解方法及存储介质: 中国, CN120317361B [P]. 2025-10-31.
- 曾娟,王尧,刘亚男,邓浩天,陈泼龙,罗炜麟,万海. 一种基于异构目标融合的精细实证溯因与声明验证方法: 中国, CN121809461B [P]. 2026-05-08.
- 罗炜麟,邓浩天,梁雪怡,万海,曾娟. 一种多级规约的软件危险行为检测方法: 中国, CN121859332B [P]. 2026-06-09.



