Yixiao HUANG

prof_pic.jpg

Kowloon, Hong Kong, China

yixiao.huang@my.cityu.edu.hk
How to pronounce?
My first name can be pronounced as “e-/ɕjɑʊ/”:

I’m Yixiao HUANG (Chinese name: 黄一笑) and I’m a first-year PhD student at UC Berkeley EECS supervised by Prof. Somayeh Sojoudi. I’m interested in establishing the theoretical foundations for language models to enhance their efficiency and reliability. My research aims to strike a balance between rigorous theoretical frameworks and thorough empirical analysis.

I received my Bachelor’s degree in Informaton Engineering in June, 2023 from Department of Electrical Engineering, City University of Hong Kong. I was fortunate to work with Prof. Rosa CHAN who patiently guided me to start my research adventure. I also had a great time at University of Michigan, working with Prof. Samet Oymak on the theoretical foundations of self-attention and hybrid models.

news

Jul 9, 2024 New paper CAT on the power of Convolution-Augmented Transformer is out!
May 3, 2024 Two papers: Sparse-PGD and From Self-Attention to Markov Models are accepted to ICML 2024. See you at Vienna.
Apr 3, 2024 I will be joining UC Berkeley EECS as a first-year PhD student in the upcoming Fall. See you at Berkeley!
Jan 10, 2024 Our paper on the implicit bias of next-token prediction has been accepted to AISTATS 2024! It’s also available on arXiv.

selected publications

* indicates equal contribution
  1. On the Power of Convolution Augmented Transformer
    Mingchen LiXuechen ZhangYixiao Huang, and Samet Oymak
    2024
  2. ICML 2024
    From Self-Attention to Markov Models: Unveiling the Dynamics of Generative Transformers
    M. Emrullah IldizYixiao Huang, Yingcong Li, Ankit Singh Rawat, and Samet Oymak
    International Conference on Machine Learning (ICML), 2024
  3. AISTATS 2024
    Mechanics of Next Token Prediction with Self-Attention
    Yingcong Li*, Yixiao Huang*M. Emrullah IldizAnkit Singh Rawat, and Samet Oymak
    International Conference on Artificial Intelligence and Statistics (AISTATS), 2024
  4. ICML 2024
    Towards Efficient Training and Evaluation of Robust Models against ł_0 Bounded Adversarial Perturbations
    Xuyang Zhong, Yixiao Huang, and Chen Liu
    International Conference on Machine Learning (ICML), 2024