高源，男，郑州大学计算机与人工智能学院，博士毕业于华中科技大学，2023年获IEEE HITC-2023 年度优秀博士奖、2025年获IEEE TCSC-2025 优秀青年学者。研究方向包括：大模型推理加速与优化、多模态深度学习、知识图谱、边缘智能等。邮箱yuangao@zzu.edu.cn。
已在IEEE TPDS/TMC/TSC/TNNLS/TII/TCSVT/BDMA/Information Fusion等中科院1区/CCF A类国际顶级期刊与ICLR/WWW/SIGIR、AAAI、ACM MM等CCF A类国际学术会议上发表论文40余篇；担任国际学术会议IEEE HPCC-2021(中国计算机学会推荐 C类国际学术会议)等会议宣传主席，IEEE DIKW-2022程序主席，IEEE Cybermatics Congress, CBD-2023执行主席，IEEE DSCI2025执行主席；担任CCF协同计算专委会/CCF嵌入式专委会执行委员、国际顶级期刊TNNLS、TCSVT、Information Fusion、TII以及国内外顶级会议NeurIPS、ICML、ACM MM、AAAI、IJCAI、WWW、ICLR审稿人。
近年来主持国家自然科学基金青年基金、国家博士后人员资助计划、博士后面上、河南省博士后科研项目资助、曙光先进计算横向课题等项目，先后参与国家重点研发计划等国家级项目3项，包括作为主要骨干成员（排名前五）参与科技部2030“新一代人工智能” 重大专项-人工智能基础模型关键技术、国家重点研发计划“网络协同制造和智能工厂 ”重点专项、国家重点研发计划“网络空间安全”重点专项等项目。
近年来发表论文
[1]RLAP-CLIP: Continual Multimodal Learning with Prototype Adaptation and Difficulty-Aware Routing. International Conference on Learning Representations (ICLR-2026)（唯一通讯，CCF A类会议长文）
[2]Learning Schema Embeddings for Service Link Prediction: A Coupled Matrix-Tensor Factorization Approach, IEEE Transactions on Services Computing, doi:10.1109/TSC.2025.3541552（唯一通讯，CCF A类期刊）
[3]Online Caching Replacement in Erasure Coding-based Edge Storage System, IEEE Transactions on Services Computing, DOI: 10.1109/TSC.2025.3619250, 2025 （唯一通讯，CCF A类期刊）
[4]Generalize to Fully Unseen Graphs: Learn Transferable Hyper-Relation Structures for Inductive Link Prediction. In Proceedings of the 32nd ACM International Conference on Multimedia, 2024:1274-1282（唯一通讯，CCF A类会议长文）
[5]From Knowledge Forgetting to Accumulation: Evolutionary Relation Path Passing for Lifelong Knowledge Graph Embedding. Proceedings of ACM SIGIR Conference-2025（唯一通讯，CCF A类会议长文）
[6]Multimodal Contextual Interactions of Entities: A Modality Circular Fusion Approach for Link Prediction. In Proceedings of the 32nd ACM International Conference on Multimedia, 2024: 8374-8382（唯一通讯，CCF A类会议长文）
[7]Tensor-Representation-Based Multiview Attributed Graph Clustering with Smooth Structure, IEEE Transactions on Neural Networks and Learning Systems, doi:10.1109/TNNLS.2025.3526590, 2025. (第一作者，中科院一区TOP, IF=10.4)
[8]Balancing Performance and Efficiency: Towards Superior Image Segmentation with Adaptive Sparse Attention. IEEE Transactions on Circuits and Systems for Video Technology, doi: 10.1109/TCSVT.2026.3651347（唯一通讯，中科院1区TOP）
[9]Multi-site brain disease identification based on tensor decomposition and personalized federated learning, Neural Networks, Doi.org/10.1016/j.neunet.2025.107987（唯一通讯，中科院1 区TOP）
[10]Referring Multi-Object Tracking with Conflict-Free Decoder Learning for Intelligent Visual Retrieval in Consumer Electronics, IEEE Transactions on Consumer Electronics, DOI: 10.1109/TCE.2026.3663002（唯一通讯，中科院 1 区TOP）
[11]Online Multi-Object Tracking Based on Record Confidence and Hierarchical Association for Cyber-Physical-Social Intelligence, Big Data Mining and Analytics，doi:10.26599/BDMA.2025.9020024, 2025. 通讯，(中科院一区, IF=7.7)
[12]Tensor Representation Based Multi-View Graph Contrastive Learning for IoE Intelligence. IEEE Internet of Things Journal, DOI:10.1109/JIOT.2024.3415612, 2024. (第一作者，中科院一区，IF=10.6)
[13]Mvtucker: Multi-view Knowledge Graphs Representation Learning Based on Tensor Tucker Model. Information Fusion, 106, 102249. 2024. 通讯，(中科院一区, IF=17.564)
[14]Jointly Low-Rank Tensor Completion for Estimating Missing Spatiotemporal Values in Logistics Systems. IEEE Transactions on Industrial Informatics, 2022, 19 (2): 1814-1822 (第一作者，中科院一区, IF=12.3)
[15]Federated Tensor Decomposition-Based Feature Extraction Approach for Industrial IoT. IEEE Transactions on Industrial Informatics, 2021,17(12): 8541 -8549. (第一作者，中科院一区, IF=12.3)
[16]Towards Multimodal Inductive Learning: Adaptively Embedding MMKG via Prototypes. WWW'25: Proceedings of the ACM Web Conference 2025. (Oral, CCF A 类会议)
[17]APT: Towards Universal Scene Graph Generation via plug-in Adaptive Prompt Tuning. The Fourteenth International Conference on Learning Representations (ICLR 2026) （CCF A类会议长文）
[18]Towards Multimodal Continual Knowledge Embedding with Modality Forgetting Modulation. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI-2025) （CCF A类会议长文）
招生信息
课题组科研经费充足，算力资源充裕！欢迎对大模型、深度学习、人工智能等方向感兴趣的硕士与本科生同学加入，请有意向同学将个人简历发送至yuangao@zzu.edu.cn。
Ø希望在本科生阶段（在读本科生）、硕士研究生阶段（本科保研、考研）、参与相关科学研究的学生，都可以联系加入课题组。
Ø团队与多所国内外科研院校保持良好合作关系，优秀者可推荐前往深造。
