Speaker：

Siheng Chen

Time：

2026.05.06 16:00-17:30

Venue：

Zhiyuan College(Kuang Piu Building)Lecture Hall, 1st floor

Abstract:

Scientific research serves as the core driver of human progress, making the enhancement of its efficiency vital for technological advancement. Exploring how artificial intelligence can accelerate scientific discovery has become a key litmus test for achieving General Artificial Intelligence (AGI). With this in mind, we propose SciMaster, a general-purpose scientific agent designed to significantly boost research efficiency by establishing a comprehensive closed-loop system encompassing searching, reading, computing, experimenting, and writing. Leveraging advanced agent technology, SciMaster demonstrates exceptional comprehensive academic capabilities and currently ranks first on the OpenAI FrontierScience leaderboard. This report will focus on the technical implementation and practical impact of SciMaster across core stages—including literature retrieval, scientific computing, and academic writing, providing a novel solution for AI-assisted research.

Bio:

Siheng Chen is an Associate Professor at the School of Artificial Intelligence, Shanghai Jiao Tong University. He received his Ph.D. from Carnegie Mellon University (CMU) and is a recipient of the National Overseas High-level Talents Program (Young Professionals). Previously, he held a position at Uber ATG within their autonomous driving division. Professor Chen has led several prestigious research initiatives, including original exploratory and general programs under the National Natural Science Foundation of China (NSFC), sub-projects of the Ministry of Science and Technology’s "Artificial Intelligence 2030" major project, and AI-specific research grants from the Shanghai Municipal Science and Technology Commission. He has published over 100 papers in premier journals and conferences—such as Nature Computational Science, Nature Communications, Cell Patterns, IEEE T-PAMI, NeurIPS, ICML, ICLR, CVPR, ICCV, and KDD—with over 10,000 citations on Google Scholar. His accolades include the IEEE Signal Processing Society Best Young Author Paper Award, the ASME Structural Health Monitoring Best Paper Runner-Up Award, the 2018 GlobalSIP Best Paper Award, and the Mitsubishi Electric Research Laboratories (MERL) President’s Award. In addition to his research, he serves as an Associate Editor for IEEE T-SIPN and has acted as an Area Chair for NeurIPS, ICML, and AAAI. His current research is centered on Agentic Science.