Prof. Xindong WU
Title: Synergizing Knowledge Graphs with Large Language Models for Bidiectional Reasoning
Bio:
Xindong Wu is the Director and Professor of the Key Laboratory of Knowledge Engineering with Big Data (the Ministry of Education of China), Hefei University of Technology, and also Chief Scientist at the CEC Data Industry Group, China. His research interests include big data analytics, data mining and knowledge engineering. He received his Bachelor’s and Master’s degrees in Computer Science from the Hefei University of Technology, China, and his Ph.D. degree in Artificial Intelligence from the University of Edinburgh, UK. He is a Foreign Member of the Russian Academy of Engineering, and a Fellow of IEEE and the AAAS (American Association for the Advancement of Science).
Abstract:
Large language models (LLMs), such as ChatGPT and GPT4o, are making new waves in the field of natural language processing and artificial intelligence, due to their emergent ability and generalizability. However, LLMs are black-box models, which often fall short of capturing and accessing factual knowledge. In contrast, Knowledge Graphs (KGs), Wikipedia, and Huapu for example, are structured knowledge models that explicitly store rich factual knowledge. KGs can enhance LLMs by providing external knowledge for inference and interpretability. Meanwhile, KGs are difficult to construct and evolve by nature, which challenges the existing methods in KGs to generate new facts and represent unseen knowledge. Therefore, it is complementary to unify LLMs and KGs together and, simultaneously, leverage their advantages. In this talk, we will discuss how KGs and LLMs can be synergized to enhance both LLMs and KGs for bidirectional reasoning driven by both data and knowledge.
Prof. Chenggang Yan
Title: Visual Perception Technology in Open Environments
Bio:
Chenggang Yan, Professor and PhD supervisor, is the Vice President of Hangzhou Dianzi University and the Director of the ‘Intelligent Information Processing’ Laboratory at Hangzhou Dianzi University. He is a Distinguished Professor of the ‘National Talent Reward Program,’ a leader of key projects under the National Key R&D Program, and a principal investigator of major projects supported by the National Natural Science Foundation of China. He is also a Young Scholar under the ‘National Talent Reward Program.’ He has long been dedicated to research in the field of intelligent information processing. In the past five years, he has published over 100 papers in international journals and conferences in this field, and his students have won the Best Paper Award at international conferences five times under his guidance. His research achievements have been recognized with the First Prize of the Natural Science Award from the China Institute of Electronics, the First Prize of the Wu Wenjun Artificial Intelligence Science and Technology Progress Award, the First Prize of the Zhejiang Provincial Science and Technology Progress Award, and the Second Prize of the National Natural Science Award.
Abstract:
In recent years, with the rapid development of large-scale vision-language models, visual perception technology has gradually expanded its application to open environments, aiming to break the limitations of closed-set object labels and improve the perception of unknown object categories. This talk will focus on the latest advancements in visual perception technology in open environments and its potential application prospects.
Prof. Tianyi Zhou
Title: Dataset Distillation and Pruning: Streamlining Machine Learning Performance
Bio:
Tianyi Zhou, Ph.D., graduated from Nanyang Technological University in Singapore and now serves as the Deputy Director and Principal Scientist at the Singapore Frontier Artificial Intelligence Center. Dr. Zhou Tianyi has led several key research and development projects in Singapore and has published more than 100 papers in core journals in fields such as machine learning, artificial intelligence, and information security (Ranked in the Chinese Academy of Sciences’ Region 1), as well as at international conferences (CCF Class A). Additionally, he serves as the deputy editor-in-chief/permanent editor of important international SCI journals such as AIJ and IEEE Transactions. Dr. Zhou has also chaired many top/important international academic conferences, such as NeurIPS, ICML, ICLR, AAAI, and IJCAI (Area Chair), and he is the vice chair for the IJCAI 2025 conference. He has won the Best Paper Award at several top/important international academic conferences, including IJCAI, ECCV, and ACML, and has delivered special reports at these conferences. He has also been recognized as one of the world’s top 2% of scientists by Stanford University.
Abstract:
In the rapidly evolving field of machine learning, “Dataset Distillation and Pruning” has emerged as a key strategy for enhancing model efficiency. Dataset distillation involves extracting essential information from extensive datasets to create refined, smaller-scale data that maintains model robustness while reducing computational burden. It can be likened to distilling knowledge from vast amounts of data.
On the other hand, dataset pruning is akin to pruning unnecessary branches from a tree. This technique involves removing redundant or minimally impactful data points, resulting in a more streamlined, faster, and resource-efficient machine learning model. By eliminating extraneous information, dataset pruning aids in constructing lean algorithms with outstanding performance and without unnecessary computational overhead.
These two approaches collectively address the challenges posed by the abundance of data in the digital age. Dataset distillation and pruning complement each other in model compression research and further optimize the entire machine learning workflow’s energy consumption, ultimately facilitating sustainable deployment of large-scale data and models on endpoints.
Prof. Guangjie Han
Title: ‘Software-Defined’ Reconstruction of Underwater Multi-Agent Networks: Research Progress and Reflections
Bio:
Guangjie Han, Ph.D., Second-level Professor/Doctoral Director, Dean of the School of Information Science and Engineering of Hehai University, IEEE/IET/AAIA Fellow. His main research directions are hydroacoustic communication and networking, industrial Internet of Things, artificial intelligence, network and security, etc. In recent years, he published 380 high-level SCI journal papers in international journals such as IEEE JSAC, IEEE TMC, IEEE TPDS, IEEE TON (including IEEE/ACM Trans. Series of journals 130+ articles), so far Google Scholar has cited 17,900+ times, and H-index is 70. He has coordinated more than 30 projects above the provincial and ministerial levels, including the national key research and development plan and the key projects of the National Natural Science Foundation, won 10 awards above the provincial and ministerial levels, and authorized 144 invention patents domesticand abroad. He has been selected into the list of scientists in the top 2% of the world for 5 consecutive years (2019-2025), and has been selected into the list of Elsevier China’s highly cited scholars for 4 consecutive years (2020-2023). At present, he is the deputy editor-in-chief of more than 10 international journals (including IEEE TII, IEEE TCCN, IEEE TVT, IEEE Systems). 333 high-level talents (second level) in Jiangsu Province, middle-aged and young experts with outstanding contributions in Jiangsu Province, Changzhou May Day Labor Medal and other honorary titles.
Abstract:
The rapid development of underwater multi-agent collaboration and swarm intelligence technologies has led to the widespread application of underwater intelligent equipment clusters, represented by underwater autonomous underwater vehicles (AUVs), in fields such as ocean engineering and marine scientific research. These advancements are expected to ultimately empower the digitalization, networking, and transparency of ocean-related activities. However, the complexity of the marine environment poses significant challenges to underwater intelligent clusters, such as communication difficulties, weak coordination, and high risks during collaborative underwater missions. These challenges have become key limiting factors in the development and application of underwater multi-agent technology.
In response, this talk introduces the concept of underwater multi-agent networks to support autonomous communication among intelligent underwater clusters. On this foundation, “software-defined networking” (SDN) technology is introduced to reconstruct the underlying architecture of underwater multi-agent networks, decoupling the network’s control and task operation planes. This allows for software-defined underwater collaborative operations, enhancing network programmability and scalability. Drawing on the research achievements of our laboratory, this report will present how SDN technology can be used to redesign data routing architectures, formation navigation frameworks, and autonomous learning decision systems in underwater multi-agent networks.
The goal of this talk is to offer a new perspective for reevaluating and building collaborative control mechanisms for underwater multi-agent systems and to lay the foundation for the construction of a “software-defined intelligent ocean,” supporting the vision of deep-sea exploration and marine advancement.