About Me

Hi, welcome to my homepage! :wave:

My name is Guanxi, I’m currently an undergraduate student in the Department of Electrical and Electronic Engineering at Imperial College London. My experience spans multiple disciplines, including computer architecture, FPGA (Field-Programmable Gate Array), and machine learning.

Currently, my research interests are efficient LLM inference, including test-time scaling, mixed-precision inference, and efficient Diffusion LLM (dLLM) inference. You can check out Experiences to see some of my experiences.


:fire: News

  • [2026.01] AdaBlock-dLLM was accepted to ICLR 2026!
  • [2025.11] FastTTS was accepted to ASPLOS 2026. Congratulations to Mark!
  • [2025.11] One invited paper on mixed precision will appear at ASP-DAC 2026!
  • [2025.09] VG-Search on Test-Time Scaling was accepted to NeurIPS 2025!
  • [2025.07] I will continue as a research intern in Dr. Hongxiang Fan’s research group, funded by the IROP Program!
  • [2025.02] I will start as a research intern in Dr. Hongxiang Fan’s research group at Imperial!
  • [2024.06] I will start as a research intern in the iDSL Lab at Imperial, supervised by Prof. Christos Bouganis. Thanks to the UROP scheme for the funding!

:game_die: Random

In my spare time, I love travelling and watching football. I’m a socio (member) of FC Barcelona, which means I’m one of the owners of the best football club in the world :satisfied: