Hi there! Thanks for stopping by :)

 I am a researcher with interests in computer vision (CV) and LLMs. I obtained M.S. in Electrical & Computer Engineering and M.S. in Data Science at the University of Michigan, Ann Arbor. During my studies, I had a wonderful internship opportunity under Adobe Research, where I was mainly mentored by Siavash Khodadadeh and Kevin Duarte and worked on diffusion model efficiency. I also had research experiences on 3D object detection at Isuzu Motors. 

Prior to my master journey, I received my B.S. from National Tsing Hua University in Taiwan, where my research lies in the field of reinforcement learning.

Currently,  I am a Machine Learning Engineer @ Google.  I work on on-device Gemini Nano model eval.

Email: hsiaoyt@umich.edu

CV/LinkedIn/Google Scholar/X

Our solution improves image quality by replacing classifier-free guidance and eliminating misalignment in unconditional scores.

Our solution for fast diffusion model sampling with efficient training and high adaptability to different fine-tuned models.

We adopted SOTA cross-modal (RGB & LiDAR) object detection algorithms for commercial trucks.

We introduce a comprehensive reinforcement learning framework for portfolio management.