About me

I am currently a Postdoctoral Researcher in Prof. Dawn Song’s group at UC Berkeley. Before that, I earned my Ph.D. in Computer Sciences from the University of Wisconsin-Madison, advised by Prof. Sharon (Yixuan) Li. My PhD research aims to pave the way to a reliable Open-world Machine Learning system, covering topics: Out-of-distribution (OOD) Detection, Open-world Representation Learning (ORL), Interpretability, etc.

My current research focuses on Large Language Models (LLMs), particularly in the following areas:

  • OOD Learnability: How can we enable LLMs to solve previously unsolvable tasks using RLVR? (Welcome contributions to this new area! GitHub stars)

  • OOD Generalization: In which aspects can inference-time reasoning be generalized in mathematical problems, programming code, and agentic tasks?

I am also interested in hallucination, interpretability, and broader aspects of trustworthy LLM research.

(Latest update: 10/03/2025)

News

  • [10/03/2025] Blog “RL Grokking Recipe” is released on RDI website.
  • [09/25/2025] Two papers are acccepted to NeurIPS 2025. See you in SD!
  • [2/26/2025] One paper is acccepted to CVPR 2025.
  • [1/22/2025] Two papers are acccepted to ICLR 2025.
  • [9/25/2024] Two papers are acccepted to NeurIPS 2024.
  • [5/1/2024] Two papers are acccepted to ICML 2024.
  • [3/13/2024] One paper is accepted to NAACL 2024.
  • [12/8/2023] One paper is accepted to AAAI 2024.
  • [9/21/2023] Two papers are accepted to NeurIPS 2023. One is spotlighted.
  • [7/19/2023] Defended my thesis “Detecting and Learning Out-of-Distribution Data in the Open world: Algorithm and Theory”. Finally Dr. Sun!
  • [4/24/2023] One paper is accepted to ICML 2023.
  • [2/27/2023] One paper is accepted to CVPR 2023.
  • [1/4/2023] One journal paper is accepted to TMLR.