Hao Li (李豪)


Ph.D. Candidate
MMLab, The Chinese University of Hong Kong

Email: haoli at link dot cuhk dot edu dot hk
Github: https://github.com/cpsxhao

I am a final year Ph.D. candidate at MMLab of the Chinese University of Hong Kong, advised by Professor Xiaogang Wang and Professor Hongsheng Li. I obtained my bachelor's degree from Tsinghua University, majoring in Electronic Engineering and double-majoring in Economics. Currently I work closely with Professor Jifeng Dai for interesting topics in multimodal foundation models and LLM-augmented agents, as well as AutoML in computer vision. I also worked with Professor Gao Huang on efficient deep learning.

I am expected to graduate in 2024, and I am actively looking for full-time research positions / internships.


Education

  • Ph.D. in Electronic Engineering, The Chinese University of Hong Kong, 2020 - Now
  • B.Eng. in Electronic Engineering, Tsinghua University, 2015 - 2020
  • B.S. in Economics (Double Major), Tsinghua University, 2016 - 2020

  • Experience

  • Research Intern, SenseTime, 11/2019 - 10/2022
  • Visiting Student, IFP Group at UIUC, 07/2019 - 09/2019
  • Research Intern, Baidu Robotics and Auto-driving Lab, 03/2019 - 06/2019
  • Research Intern, Microsoft Research Asia, 08/2018 - 01/2019

  • Selected Publications & Preprints (Google Scholar)

    Auto MC-Reward: Automated Dense Reward Design with Large Language Models for Minecraft. [paper]
    Hao Li*, Xue Yang*, Zhaokai Wang*, Xizhou Zhu, Jie Zhou, Yu Qiao, Xiaogang Wang, Hongsheng Li, Lewei Lu, Jifeng Dai.
    International Conference on Computer Vision and Pattern Recognition (CVPR) 2024.

    Uni-Perceiver v2: A Generalist Model for Large-Scale Vision and Vision-Language Tasks. [paper] [poster] [blog(CN)]
    Hao Li*, Jinguo Zhu*, Xiaohu Jiang*, Xizhou Zhu, Hongsheng Li, Chun Yuan, Xiaohua Wang, Yu Qiao, Xiaogang Wang, Wenhai Wang, Jifeng Dai.
    International Conference on Computer Vision and Pattern Recognition (CVPR) 2023. Highlight, 2.5% acceptance rate

    Uni-Perceiver: Pre-training Unified Architecture for Generic Perception for Zero-shot and Few-shot Tasks. [paper] [blog(CN)] [code]
    Xizhou Zhu*, Jinguo Zhu*, Hao Li*, Xiaoshi Wu*, Xiaogang Wang, Hongsheng Li, Xiaohua Wang, Jifeng Dai.
    International Conference on Computer Vision and Pattern Recognition (CVPR) 2022.

    AutoLoss-Zero: Searching Loss Functions from Scratch for Generic Tasks. [paper] [poster] [blog(CN)]
    Hao Li*, Tianwen Fu*, Jifeng Dai, Hongsheng Li, Gao Huang, Xizhou Zhu.
    International Conference on Computer Vision and Pattern Recognition (CVPR) 2022.

    Auto Seg-Loss: Searching Metric Surrogates for Semantic Segmentation. [paper] [blog(CN)] [code]
    Hao Li*, Chenxin Tao*, Xizhou Zhu, Xiaogang Wang, Gao Huang, Jifeng Dai.
    International Conference on Learning Representations (ICLR) 2021.

    Improved Techniques for Training Adaptive Deep Networks. [paper] [poster] [code]
    Hao Li*, Hong Zhang*, Xiaojuan Qi, Ruigang Yang, Gao Huang.
    International Conference on Computer Vision (ICCV) 2019.

    * Equal Contribution.