Jack (Hao) Bai

haob2 AT illinois DOT edu

pic.jpg

Hi there! I’m Jack, a second-year thesis-based master student affiliated with UIUC CS, advised by Prof. Nan Jiang. I’m also a visiting scholar at BAIR, advised by Sergey Levine.

Recently, I focus my research on building intelligent agents with foundation model and reinforcement leanring (RL). The methodology usually integrates principled representation learning and RL algorithms.

I received my dual undergrad degree from UIUC and Zhejiang University. During those wonderful years, I was lucky enough to work with Yi Ma @ BAIR, Chengxiang Zhai @ UIUC, and Shilin He @ Microsoft Research.

A public up-to-date resume can be found here.

News

Oct 23, 2024 My first representation learning paper CRATE-LM has been released! This work was done when I visited BAIR, advised by Prof. Yi Ma.
Sep 28, 2024 My two proud works DigiRL (first author) and RL4VLM (second author) have been accepted to NeurIPS 2024! 🎉🎉🎉
Jun 20, 2024 My first first-authored paper DigiRL is released and selected as Oral at Foundation Models in the Wild Workshop @ ICML’2024. This work is gracefully under guidance of Prof. Aviral Kumar and Prof. Sergey Levine.

Latest Posts

Selected Publications

  1. Preprint
    Improving Neuron-level Interpretability with White-box Language Models
    Hao Bai , and Yi Ma
    Oct 2024
  2. NeurIPS 2024 Oral @ ICML WS
    DigiRL: Training In-The-Wild Device-Control Agents with Autonomous Reinforcement Learning
    Hao Bai , Yifei Zhou, Jiayi Pan, Mert Cemri, Alane Suhr, Sergey Levine, and Aviral Kumar
    Jun 2024
  3. NeurIPS 2024
    Fine-Tuning Large Vision-Language Models as Decision-Making Agents via Reinforcement Learning
    Yuexiang Zhai,  Hao Bai , Zipeng Lin, Jiayi Pan, Shengbang Tong, Yifei Zhou, Alane Suhr, Saining Xie, Yann LeCun, Yi Ma, and Sergey Levine
    May 2024
  4. JMLR
    White-Box Transformers via Sparse Rate Reduction: Compression Is All There Is?
    Yaodong Yu, Sam Buchanan, Druv Pai, Tianzhe Chu, Ziyang Wu, Shengbang Tong,  Hao Bai , Yuexiang Zhai, Benjamin D Haeffele, and Yi Ma
    Apr 2024
  5. WSDM’24
    CharmBana: Progressive Responses with Real-Time Internet Search for Knowledge-Powered Conversations
    Revanth Gangi Reddy, Sharath Chandra,  Hao Bai , Wentao Yao,  ..., and Chengxiang Zhai
    Feb 2024
  6. EMNLP’23
    Social Commonsense-Guided Search Query Generation for Open-Domain Knowledge-Powered Conversations
    Revanth Reddy,  Hao Bai , Wentao Yao, Sharath Chandra Etagi Suresh, Heng Ji, and ChengXiang Zhai
    Oct 2023