Haotian Hu

Georgia Institute of Technology

prof_pic.jpg

Big fan of Leagues of Legend

Stardew Valley & Terraria

This is Haotian Hu 「胡皓天」. I’m currently an first year Machine Learning Ph.D. student at Georgia Institute of Technology, working with Prof.Faramarz Fekri. Previously, I recieved my bachelor degree from University of Science and Technology of China, School of the Gifted Young.

My current research focuses on building efficient and reliable machine learning systems, specifically in the following topics: (i) Hypothesis generation with Large Language Models; (ii) Large Language Models Towards Reasoning; (iii) Deep Learning Theory;

I was fortunate to cooperate with Zhenyu Zhang@UT-Austin (Atlas Wang - VITA), Tianjin Huang@UOE, Yufei Ding@UCSD, Sitao Huang@UCI and Kai Zhong@UESTC.

You can find me through Google Scholar and Gmail.

selected publications

  1. Stable-SPAM: How to Train in 4-Bit More Stably than 16-Bit Adam
    Tianjin Huang*, Haotian Hu*, Zhenyu Zhang*, and 8 more authors
    arXiv preprint arXiv:2502.17055, 2024
  2. Optimizing Quantum Communication for Quantum Data Centers with Reconfigurable Networks
    Hezi Zhang, Yiran Xu, Haotian Hu, and 6 more authors
    ISCA 2025 to appear, 2024
  3. Optimizing High-Level Synthesis Designs with Retrieval-Augmented Large Language Models
    Haocheng Xu, Haotian Hu, and Sitao Huang
    In 2024 IEEE LLM Aided Design Workshop (LAD), 2024
  4. Ambiguity Function Shaping via Manifold Optimization Embedding With Momentum
    Haotian Hu, Kai Zhong, Cunhua Pan, and 1 more author
    IEEE Communications Letters, 2023