Guangsheng Bao

PhD candidate @ Westlake University, Natural Language Processing.

prof_pic.jpg

Hi, thanks for stopping by!

I am Guangsheng Bao, currently a Ph.D. candidate at Westlake University and Zhejiang University, where I have the privilege of being advised by Prof. Yue Zhang. My primary research interests lie in the field of trustworthy NLP, encompassing topics like model generalization, the creation of explainable and controllable NLP systems, ensuring the safe application of LLMs, and understanding the reasoning capabilities of LLMs. I have disseminated my research outcomes across a range of publications presented at esteemed conferences including ACL, ICLR, EMNLP, TASLP, and AAAI.

In addition to my research, I’ve been actively involved in the academic community as a reviewer for conferences such as ACL, EMNLP, ICML, NIPS, TACL, COLING, and more. I’ve also had the honor of serving as an area chair for COLING 2022.

Before diving into the world of scientific research, I gained valuable experience at Alibaba, contributing to customer service robot systems, and at Microsoft (China), where I worked on Cortana. Feel free to reach out if you would like to chat about any of these topics or if you have any questions!

Email: baoguangsheng@westlake.edu.cn

Address: No. 600 Dunyu Road, Sandun Town, Xihu District, Hangzhou, Zhejiang PR China.

news

Jan 23, 2025 :fire: Three papers are accepted by ICLR 2025.
Dec 2, 2024 :fire: One paper is accepted by COLING 2025.
Apr 27, 2024 Demo for Fast-DetectGPT is online!
Jan 17, 2024 Two papers are accepted by ICLR 2024.
Oct 8, 2023 Two papers are accepted by EMNLP 2023.

selected publications

  1. ICLR 2024
    Fast-DetectGPT: Efficient Zero-Shot Detection of Machine-Generated Text via Conditional Probability Curvature
    Guangsheng Bao, Yanbin Zhao, Zhiyang Teng, and 2 more authors
    The Twelfth International Conference on Learning Representations, 2024
  2. COLING 2025
    How Likely Do LLMs with CoT Mimic Human Reasoning?
    Guangsheng Bao, Hongbo Zhang, Linyi Yang, and 2 more authors
    In Proceedings of the 31st International Conference on Computational Linguistics, 2025
  3. ICLR 2025
    Glimpse: Enabling White-Box Methods to Use Proprietary Models for Zero-Shot LLM-Generated Text Detection
    Guangsheng Bao, Yanbin Zhao, Juncai He, and 1 more author
    The Thirteenth International Conference on Learning Representations, 2025
2,291 Total Pageviews