LI Yikang's Homepage

Associate Professor (Tenure-track) @ SII (上海创智学院)
Quantitative Researcher & Partner @ FFT Investment
Room 611, 699 Huafa Road, Xuhui District, Shanghai
lyk [at] sii.edu.cn
Google Scholar | GitHub | LinkedIn | Resume

I bridge the gap between AI research and real-world impact. Currently, I lead DELTA Lab at Shanghai Innovation Institute, focusing on collective intelligence and AI-powered event perception & forecasting systems. In parallel, I serve as a Partner at FFT Investment, where I lead AI and data teams to improve global asset valuation using advanced ML models.

My career spans academia, industry R&D, and venture capital — a unique trajectory that shapes my research philosophy: data-driven, practice-oriented, and impact-focused.

Research Interests: Multi-modal Fusion, Continual Learning, Event Perception & Forecasting, and Data-driven Systems.

📢 Recruiting! Looking for self-motivated PhD students, research interns, and engineers to join DELTA Lab. If you are passionate about using AI and data to solve real-world problems, please reach out with your CV. See Prospective Students for details.

Prior Experience:

2022 - 2024 · VP in AI · IDG Capital

Invested in Embodied AI & LLM: Galaxea AI, Step AI, MiniMax, etc.

2021 - 2023 · Research Scientist & PI · Shanghai AI Lab

Founded ADLab (35+ researchers); #1 on Waymo 3D Detection.

2019 - 2022 · Deputy Director of R&D · SenseTime

Led L4 autonomous driving product team (40+ members).

Education:

2016 - 2020 · Ph.D. in CV & ML · CUHK-MMLab

HK PhD Fellowship & Microsoft Research PhD Fellowship

2011 - 2016 · B.S. in EE & Economics (dual) · Tsinghua

GPA: 92/100, Ranking: 1/18, National Scholarship

News

Dec 2025 · Recruiting PhD students & interns for DELTA Lab. Details NEW!

Jul 2025 · Joined SII as Associate Professor (Tenure-track).

Jul 2023 · 4 Papers are accepted by ICCV-23.

Feb 2023 · 6 Papers are accepted by CVPR-23.

Jan 2023 · 2 Papers are accepted by ICRA-23, Sensor Calibration * 1 and Sensor Placement * 1.

Nov 2022 · One paper is accepted by AAAI-23, Instance Segmentation for Autonomous Driving.