微软亚洲研究院 ACE Talk 系列讲座旨在邀请杰出的学术新星分享科研成果，为学生与研究员提供相互交流学习与洞悉前沿动态的平台。

第九期 ACE Talk，我们特别邀请到来自北卡罗来纳州立大学计算机系的助理教授胥栋宽为我们带来以“Sculpting the Future of Collective Growth in Collaborative AI: Gentopia.AI Meets ReWOO”为主题的报告，介绍他借助Gentopia推动人工智能代理之间的协同互动的研究工作。欢迎大家报名！

讲座信息

时间：9 月 26 日（周二）10：00-11：30

地点：线上

日程：

• 嘉宾报告（10:00-11:00）

• Q&A（11:00-11:30）

报名方式

欢迎扫描下方二维码填写报名问卷，报名成功后将收到邮件通知，邮件中将提供讲座 Teams 线上会议链接。

报名截止时间：9 月 24 日（周日）12：00

报名链接：

https://jinshuju.net/f/yvN0X6

嘉宾介绍

Dongkuan (DK) Xu is an Assistant Professor in the CS Department at NC State and leads the NCSU Generative Intelligent Computing Lab. His research is fundamentally grounded in exploring and advancing Landed Generative AI, with particular emphasis on studying the autonomy of intelligent agents (tools, chains, tasks), decision/reasoning reliability (alignment, uncertainty, adaptability, robustness), and resource efficiency (data utilization, computation, parameter optimization) in Generative AI (ChatGPT, GPT-X, Diffusion Models) Systems. DK’s research has been published repeatedly in top conferences and journals in AI, NLP, CV, Hardware, and other fields. He serves as the Column Editor for ACM SIGAI Newsletter, chairs The First Workshop on DL-Hardware Co-Design for AI Acceleration with AAAI 2023, and co-chairs The Resource-Efficient Learning for Knowledge Discovery Workshop with KDD 2023. He has served as session chair for New Deep Learning Architectures, and for Scalable & Trustable AI at KDD 2022, and has served as a (senior) PC member and regular reviewer for over 30 major conferences and 16 journals. In addition, he has launched the Machine Learning Algorithms & Natural Language Processing community. DK also has extensive research experience in industry. He has been collaborating with Microsoft Research, Google Research, Moffett AI, and NEC labs America, and holds 10 US patents/applications. DK's long-term research goal is to liberate AI productivity and democratize its application to serve a broader range of populations and real-world applications, equally, sustainably, and responsibly.

报告简介

Sculpting the Future of Collective Growth in Collaborative AI: Gentopia.AI Meets ReWOO

As the scope and complexity of tasks that AI can address continue to grow, the need for collaborative multi-agent systems becomes increasingly crucial. Yet, creating an environment where specialized and generalist agents can efficiently collaborate has remained a considerable challenge. This talk unveils Gentopia.AI (https://github.com/Gentopia-AI/Gentopia), a pioneering platform designed to facilitate agent collaboration for collective growth across diverse scenarios. We introduce the concept of Gentopia.AI as an enabler for synergistic interactions among various AI agents and delve into its composition and collaboration features. To illustrate the efficacy of Gentopia.AI, we present ReWOO (https://github.com/billxbf/ReWOO)—a groundbreaking methodology that drastically improves token efficiency and accuracy in augmented language models. Through ReWOO, we showcase how Gentopia.AI fosters environments conducive to innovations that not only reduce redundancy but also scale effectively. This talk aims to inspire a new wave of research and development focused on maximizing the collective growth of intelligent systems in a collaborative ecosystem.

主持人简介

Jianxun Lian is now a senior researcher at Microsoft Research Asia. He received his Ph.D. degree from University of Science and Technology of China in 2018. His research interests include recommender systems and AI agents. He has published 30+ academic papers on top-tier international conferences such as KDD, ACL, WWW, and SIGIR. He actively contributes as a program committee member for prestigious conferences such as AAAI, WWW, and IJCAI.

关于ACE Talk

ACE Talk（Advance, Creativity and Empowerment）是由微软亚洲研究院举办的系列讲座，旨在邀请杰出的学术新星分享科研成果，为学生与研究员提供相互交流学习与洞悉前沿动态的平台。

ACE Talk Series 于 2023 年 3 月开启，迄今已成功邀请哥伦比亚大学博士钟宇宏、杜克大学助理教授 Neil Gong、香港大学助理教授余涛、南加州大学助理教授赵洁玉、麻省理工学院副教授韩松、斯坦福大学助理教授杨笛一、佐治亚理工学院助理教授 Srijan Kumar、卡耐基梅隆大学副教授朱海一等嘉宾，分享内容涉及大模型安全、自然语言界面、LLM 量化、模型社会责任等主题。未来 ACE Talk 还将继续专注于前沿研究分享，为更多领域内研究者提供多元互通的学术交流空间。