I am currently a Master’s from Department of Computing of The Hong Kong Polytechnic University. Prior to this, I received my Bachelor’s degree in Mathematics and Applied Mathematics from Hefei University of Technology (HFUT) in 2024.

My research interests primarily focus on Efficient LLM Inference, Multimodal Document Understanding, and Multi-Agent Systems. I am currently working as a Multimodal Large Model Algorithm Engineer Intern at Shanghai Artificial Intelligence Laboratory (OpenDataLab).

You can find all of my publications on my Google Scholar profile.

You can find my CV here: Jiayu’s Curriculum Vitae.

🔥 News

• [2026.05] Our paper “HybridFlow: Adaptive Task Scheduling for Fast and Token-Efficient LLM Inference in Edge-Cloud Collaboration” was accepted by ICML 2026!
• [2026.05] Our paper “AscendKernelGen: LLM-Driven Kernel Generation for NPUs” was accepted by ACL 2026 Findings!
• [2025.12] I joined the Shanghai AI Laboratory (OpenDataLab) as a Multimodal LLM Algorithm Engineer Intern.
• [2025.08] I joined Huawei (Computing Product Line) as an AI Engineer Intern.

Hobbies

I particularly enjoy photography (mainly landscape photography), instrumental music (especially the background music from Genshin Impact created by HOYO-MiX and Chen Zhiyi), and comedy (such as crosstalk and stand-up).