2026 Conference on Large Models and Intelligent Agent Technologies (LMIAT 2026) Successfully Concluded in Hangzhou

From July 17 to 19, 2026, the 2026 Conference on Large Models and Intelligent Agent Technologies (LMIAT 2026) was successfully held in Hangzhou. The conference was hosted by Zhejiang Shuren University, organized by the College of Information Science and Technology of Zhejiang Shuren University, and co-organized by the Zhejiang Federation of Artificial Intelligence and the Zhejiang Society for Health Products Safety. The event brought together experts and scholars from home and abroad to engage in in-depth exchanges and discussions on frontier topics including foundational theories, core technologies, and cross-scenario innovative applications of large models and intelligent agents.

group photo

The conference officially opened on the morning of July 18 at Conference Room 230, Cha Ji Min Building, Zhejiang Shuren University. The opening ceremony was hosted by Professor Fengjun Hu, Dean of the College of Information Science and Technology of Zhejiang Shuren University. Professor Jun Chen, Vice President of Zhejiang Shuren University, delivered a welcome address, extending a warm welcome to all participating experts and scholars. He noted that the conference focuses on core topics including efficient training of large models, multimodal fusion, cognitive architectures, and multi-agent collaboration, aiming to drive paradigm shifts in autonomous decision-making and human-machine collaboration, and to accelerate the translation of research breakthroughs into industrial impact. A group photo of all attendees was taken afterwards.

Welcome Remark: Professor Jun Chen, Vice President of Zhejiang Shuren University, China

Host: Prof. Fengjun Hu, Zhejiang Shuren University, China

In the morning session, four distinguished scholars delivered insightful presentations at the cutting edge of their fields. First, Professor Weijia Jia, Member of the US National Academy of Artificial Intelligence and IEEE Fellow from Beijing Normal University (Zhuhai), opened with a talk entitled "Open-Loop Agent Tabular QA," focusing on the challenges faced by large language models in open-domain tabular question answering. Professor Chenguang Yang, IEEE Fellow from The Hong Kong Polytechnic University, presented on "Human-like Robot Control, Skill Learning and Human Robot Interaction," addressing the bottlenecks of force-position coupling and skill learning in physical human-robot interaction, and sharing his latest research achievements in human-like compliant control, multi-modal skill primitive systems, and adaptive generalization. Professor Haoping Wang, IEEE Senior Member from Nanjing University of Science and Technology, delivered a report entitled "Ultra-Local Model and Intelligent Technique based Assisted-As-Needed Control for Rehabilitation Exoskeleton Robotic Systems," offering an in-depth analysis of the application of intelligent technologies in rehabilitation exoskeleton robotics. Dr. Zelin Zang from the Hong Kong Institute of Innovation, Chinese Academy of Sciences, gave a talk on "Logic-Tree White-Box Multi-Agent Systems for Explainable LLM Reasoning and Applications in Medicine," introducing white-box multi-agent systems for explainable large language model reasoning and their application prospects in the medical field.

Prof. Weijia Jia, Member of US National Academy of Artificial Intelligence, IEEE Fellow, Beijing Normal University, Zhuhai, China Speech Title:Open-Loop Agent Tabular QA
Prof. Chenguang Yang, IEEE Fellow, The Hong Kong Polytechnic University, China Speech Title:Humanoid Control, Skill Learning, and Human-Robot Collaboration of Robots
Prof. Haoping Wang, IEEE Senior Member, Nanjing University of Science and Technology, China Speech Title:Ultra-Local Model and Intelligent Technique based Assisted-As-Needed Control for Rehabilitation Exoskeleton Robotic Systems
Research Zelin Zang，Hong Kong Institute of Innovation, Chinese Academy of Sciences, China Speech Title:Logic-Tree White-Box Multi-Agent Systems for Explainable LLM Reasoning and Applications in Medicine

The afternoon session was hosted by Associate Professor Guoyong Dai, Associate Dean of the College of Information Science and Technology at Zhejiang Shuren University, who presided over the invited speech and oral presentation sessions. Dr. Hangyao Tu from Zhejiang Shuren University then shared his research on "Application Research on Parametric Design Intent Understanding of Industrial CAD Large Models Based on Strategy Fine-Tuning," showcasing cutting-edge explorations in parametric design intent understanding of industrial CAD large models. Senior Engineer Zhaoyang He, CTO of AscendGrace Co., Ltd., delivered a talk from an industrial perspective entitled "From Machine Language Models (MLM) to Machine Language Intelligence: A Software Intelligence Revolution in the Era of Foundation Models," exploring the software intelligence revolution in the era of foundation models.

Dr. Hangyao Tu, Zhejiang Shuren University, China Speech Title:Application Research on Parametric Design Intent Understanding of Industrial CAD Large Models Based on Strategy Fine-Tuning
Zhaoyang He, Senior Engineer, CTO, AscendGrace Co., Ltd., China Speech Title:From Machine Language Models (MLM) to Machine Language Intelligence: A Software Intelligence Revolution in the Era of Foundation Models

In the oral presentation session, three young scholars—Li Song and Hui Xie from Zhejiang Shuren University, and Binqi Shen from Northwestern University, USA—presented their research on cutting-edge topics including lightweight graph convolutional reinforcement learning, bioinformatics workflow agents, and cost-performance optimization in large language model context management, fully demonstrating the innovative vitality of the new generation of researchers in the fields of intelligent agents and large models.

Li Song, Zhejiang Shuren University, China Speech Title:Lightweight Graph Convolutional Reinforcement Learning via Structurally Sparse Multi-scale Dilation Attention	Hui Xie, Zhejiang Shuren University, China Speech Title:BioFlowAgent: A Registry-Constrained LLM Agent Separating Planning from Execution for Reproducible Bioinformatics Workflows
Binqi Shen, Northwestern University,  America Speech Title:The Efficiency Frontier: A Unified Framework for Cost-Performance Optimization in LLM Context Management

Finally, participating experts and scholars visited the Institute of Artificial Intelligence for Traditional Chinese Medicine at Zhejiang Shuren University for an academic tour. They toured the institute and engaged in in-depth discussions on the application of large models and intelligent agent technologies in healthcare and other domains.

With the collective efforts of all parties involved, LMIAT 2026 achieved great success, fully showcasing the latest research outcomes and cutting-edge developments in the field of large models and intelligent agent technologies, and facilitating the exchange and collision of academic ideas. Looking ahead, we look forward to welcoming more outstanding scholars, experts, and young talents to join us in advancing innovation in large models and intelligent agent technologies, breaking through technological bottlenecks, and contributing to technological progress and industrial upgrading.

We look forward to meeting you again at LMIAT 2027!