Haoyang Zheng

Haoyang Zheng

Ph.D. Candidate
Purdue University

About Me

My name is Haoyang (pronounced as “Hau Yeung”); I was born in Zhejiang and grew up in Shenzhen. I am currently a PhD candidate at the School of Mechanical Engineering, Purdue University, advised by Prof. Guang Lin.

I am currently a Research Intern at Google (Mountain View, CA), focusing on LLM pre-training and post-training. Previously, I was part of the ACE Intern Program / RedStar at Xiaohongshu Inc.’s Humane Intelligence Lab (hi lab) (Beijing, China), where I worked on multimodal LLMs.

I study Generative Modeling and Reinforcement Learning (RL) from a probabilistic modeling perspective. My interests include Large Language Models (LLMs), continuous and discrete diffusion, RL, and sampling/optimization methods that improve latency and robustness. My goal is to build fast and reliable LLMs.

I am on the 2026/2027 job market for industrial positions in LLMs and multimodal AI, with a preference for roles based in the U.S., China (Hong Kong, Shenzhen, Shanghai, Hangzhou), or Singapore. Feel free to reach out via email or wechat.

Research Interests

  • Large Language Models (Pre-training & Post-training)
  • Multimodal Generative Models (VLM, Image Editing, Post-training)
  • Generative Models (Continuous/Discrete Diffusion Models, Diffusion Language Models, Autoregressive Models)

News

Services

  • Journal Reviewer: TPAMI, TMM, TNNLS, TCYB, TMLR
  • Conference Reviewer: ICLR, ICML, NeurIPS, AISTATS, AAAI, KDD

Contact

  • Email: lastname+528 at purdue dot edu; firstname+lastname+ai at google dot com
  • WeChat: zhy