Program

Keynote

Dr. Jie Li is a Chair Professor in Computer Science and Engineering, School of Electronic Information and Electrical Engineering, Director of Intelligent Big Data System (iBDSys) Lab., Director of SJTU Blockchain Research Centre, Adjunct Professor of Zhiyuan College with Zhiyuan Honors Program and SJTU Paris Elite Institute of Technology (SPEIT), Shanghai Jiao Tong University (SJTU), Shanghai, China. He serves as the Chief Scientist, Shanghai Key Lab. of Trusted Data Circulation and Governance and Web3. He is an IEEE Fellow, AAIA Fellow, and a Foreign Fellow of the Engineering Academy of Japan (EAJ). He is on the advisory board of Shanghai Frontiers Science Center for Artificial Intelligence and Deep Learning, NYU Shanghai – New York University. He was a full Professor in Department of Computer Science, University of Tsukuba, Japan. He was a visiting Professor in Yale University, USA, Inria Sophia Antipolis and Inria Grenoble-Rhone-Aples, France, during the sabbatical year in September 2014 through August 2015. He has been a visiting Professor/Adjunct Professor of The Chinese University of Hong Kong (CUHK), The Hong Kong Polytechnical University (PolyU), City University of Hong Kong, Tsinghua University, and University of Science and Technology of China (USTC). His current research interests are in Big Data, AI, Blockchain, network systems and security, and smart city. His research has been continuously supported by many national and industrial research grants. He is a Fellow of IEEE, a senior member of ACM, and a member of CCF and IPSJ. He is the founding chair of the IEEE ComSoc Technical Committee on Big Data (TCBD) and the co-chair of IEEE Big Data Community.

Prof. Weijia Jia now is servingas the Director of AI and Future Networking Research Institute of Beijing Normal University (BNU, Zhuhai), Guangdong, China. He has served as the VP for Research atBNBU (UIC). Prior joing BNU, he served as the Deputy Director of StateKay Laboratory of Internet of Things for Smart City atthe University of Macau(2018-2020) and was Zhiyuan Chair Professor at Shanghai Jiaotong University (2014-2018). He worked in CityUniversity of Hong Kong as a professor (1995-2103). His contributions have been reconganized as intelligent edge computing and LLM applications, optimal network routing and deployment; vertex cover; anycast and QoS routing, and sensors networking; knowledge relation extractions and NLP. He has over 600 publications in the prestige international journals/conferencesand research books and book chapters. He has received many science-tech awards and served as area editor for various prestige international journals, chair and PC member/keynote speaker for top international conferences. He is the Fellowof IEEE and the Distinguished Member of CCF.

Speech Title: BridgingEdges andLLMs

Abstract: Large Language Models (LLMs) are widely usedacross various domains, but deploying them in cloud datacenters often leads to significant response delays and high costs,undermining Quality of Service (QoS) at the network edge.Although caching LLM request results at the edge using vectordatabases can greatly reduce response times and costs for similarrequests, this approach has been overlooked in prior research. Toaddress this, we propose a novel Vector database-assisted cloud-Edge collaborative LLM QoS Optimization (VELO) frameworkthat caches LLM request results at the edge using vectordatabases, thereby reducing response times for subsequent similar requests. Unlike methods that modify LLMs directly, VELOleaves the LLM’s internal structure intact and is applicableto various LLMs. Building on VELO, we formulate the QoSoptimization problem as a Markov Decision Process (MDP)and design an algorithm based on Multi-Agent ReinforcementLearning (MARL). Our algorithm employs a diffusion-basedpolicy network to extract the LLM request features, determiningwhether to request the LLM in the cloud or retrieve results fromthe edge’s vector database. Implemented in a real edge system,our experimental results demonstrate that VELO significantlyenhances user satisfaction by simultaneously reducing delaysand resource consumption for edge users of LLMs. Our DLRSalgorithm improves performance by 15.0% on average for similarrequests and by 14.6% for new requests compared to thebaselines.

Professor Yang Wangdong is the Vice President of Changsha University and a recipient of China’s National High-Level Talent Program. His primary research area is high-performance computing, and he has long been engaged in the design and application of parallel algorithms on domestic supercomputing platforms and homegrown processors such as Feiteng, Kunpeng, Hygon, Ascend, and Jingjiawei GPUs. He has led the development of a series of foundational numerical algorithm libraries and performance-optimization tools. Professor Yang’s work has been honored with the Second Prize of the National Science and Technology Progress Award, the First and Second Prizes of the Hunan Provincial Natural Science Award, and two Huawei Spark Awards. He has served as principal investigator on over twenty projects, including key and general programs funded by the National Natural Science Foundation of China, National Key R&D Program initiatives, Hunan Province’s key R&D plans, and “Jiebang Guashua” (leadership pledge) projects.

Speech Title: Intelligent Solver for Data-driven Large-scale Linear Equation Systems

Solving systems of linear equations is the core technology of industrial software computing engines. As problem sizes grow and application scenarios diversify, traditional numerical algorithms and today’s end-to-end AI solvers struggle to balance reliability, solution accuracy, performance, and generalization. Integrating AI techniques to accelerate parameter-space and solution-space search in classical numerical methods can both enhance generalization and preserve interpretability and precision. This project targets three key scientific challenges—the sourcing of training data, the generalization and efficiency of solution methods, and the quality of preprocessing matrix generation—and carries out four research tasks: feature analysis and construction for linear systems; AI-driven data augmentation for sparse matrices; development and optimization of multimodal, multi-expert selection models; and intelligent preprocessing optimization. We will build a comprehensive standard dataset of sparse matrices; develop an adaptive solver that fuses traditional numerical algorithms with AI; and create an end-to-end high-efficiency preprocessing pipeline. Leveraging China’s rich application-scenario data and AI technology to leap forward in domain prior knowledge will help Chinese industrial software achieve a strategic curve-overtaking advantage.