报告题目: Implementing Hardware for Sustainable Intelligence: From Spin-Orbitronics to ICT with Edge Computing & Agents

主讲人： 邢国忠，中国科学院微电子研究所研究员

报告时间：2026年6月8日（星期一）15:00-16:00

报告地点：厦门大学翔安校区西部片区信息学院3号楼103会议室

报告摘要：

The booming AI, autonomous agents and IoT have pushed ICT energy consumption to an unsustainable level, caused by the von Neumann bottleneck. The speaker introduces a hardware solution for green computing. Recent progress in spin-orbit torque materials and magnetic tunnel junctions delivers ultra-low-power devices. Spintronic technologies are adopted to build neuromorphic hardware simulating synapses and neurons, supporting on-device AI and greatly reducing carbon emissions. A self-developed SOT-MRAM chip achieves accurate keyword spotting, proving the viability of spintronic in-memory computing. He also envisions combining these hardware with edge systems to build high-performance, eco-friendly ICT infrastructure and discusses potential academic collaborations.

报告人简介：

Dr. XING is a professor at the Institute of Microelectronics and the University of Chinese Academy of Sciences. He has been listed among the World’s Top 2% Scientists for five consecutive years. His research focuses on in-memory computing and neuromorphic hardware.

He has authored over 100 papers with more than 11,300 citations, and holds numerous Chinese and US patents. He has led multiple national key R&D programs. With rich work experience in renowned semiconductor firms and overseas universities, he earned two PhD degrees in physics.

Awarded for teaching and mentoring, he devotes himself to cultivating talents in green computing and intelligent systems.

邀请人：计算机科学与技术系 向乔教授