Hi! My name is Leyi Pan (潘乐怡). I am a first-year master’s student at the School of Software in Tsinghua University, where I am advised by Prof. Lijie Wen. Before that, I received my B.Eng. degree from the School of Software in Tsinghua University in 2024.

My research focuses on two main areas:

• AI Safety with an emphasis on watermark for large language models.
• Post-training methods to enhance reasoning capabilities of multimodal large language models.

I am currently serving as a Research Intern at Zhipu AI, where I am working under the supervision of Dr. Shiyu Huang.

Previously, I was serving as a Research Assistant at CUHK MISC Lab, working under the supervision of Prof. Irwin King (ACM Fellow, IEEE Fellow).

Education

Tsinghua University, China

Master's Student in Software Engineering

Sept. 2024 - June 2027 (Expected)

Tsinghua University, China

B.E. in Software Engineering

Sept. 2020 - June 2024

Work Experience

Zhipu AI, China

Large Multimodal Model Research Intern

Oct. 2024 - Present

The Chinese University of Hong Kong (CUHK)

Research Assistant

Jun. 2024 - Aug. 2024

KuaiShou Technology, China

Software Engineer Intern

Jun. 2023 - Aug. 2023

Research Paper

Can LLM Watermarks Robustly Prevent Unauthorized Knowledge Distillation?

Leyi Pan, Aiwei Liu, Shiyu Huang, Yijian Lu, Xuming Hu, Lijie Wen, Irwin King, Philip S. Yu

Proceedings of ACL 2025

[Paper] [Code]

WaterSeeker: Pioneering Efficient Detection of Watermarked Segments in Large Documents

Leyi Pan, Aiwei Liu, Yijian Lu, Zitian Gao, Yichen Di, Shiyu Huang, Lijie Wen, Irwin King, Philip S. Yu

Findings of NAACL 2025, AAAI 2025 Workshop PDLM (Oral)

[Paper] [Code]

MarkLLM: An Open-Source Toolkit for LLM Watermarking

Leyi Pan, Aiwei Liu, Zhiwei He, Zitian Gao, Xuandong Zhao, Yijian Lu, Binglin Zhou, Shuliang Liu, Xuming Hu, Lijie Wen, Irwin King, Philip S. Yu

Proceedings of EMNLP 2024 Demo

[Paper] [Code] [Google Colab] [机器之心]

A Survey of Text Watermarking in the Era of Large Language Models

Aiwei Liu*, Leyi Pan*, Yijian Lu, Jingjing Li, Xuming Hu, Xi Zhang, Lijie Wen, Irwin King, Hui Xiong, Philip S. Yu

ACM Computing Surveys

[Paper] [Home] [机器之心] [Twitter]

Can Watermarked LLMs be Identified by Users via Crafted Prompts?

Aiwei Liu, Sheng Guan, Yiming Liu, Leyi Pan, Yifei Zhang, Liancheng Fang, Lijie Wen, Philip S. Yu, Xuming Hu

Proceedings of ICLR 2025

[Paper] [Code]

A Semantic Invariant Robust Watermark for Large Language Models

Aiwei Liu, Leyi Pan, Xuming Hu, Shiao Meng, Lijie Wen

Proceedings of ICLR 2024

[Paper] [Code]

An Unforgeable Publicly Verifiable Watermark for Large Language Models

Aiwei Liu, Leyi Pan, Xuming Hu, Shuang Li, Lijie Wen, Irwin King, Philip S. Yu

Proceedings of ICLR 2024

[Paper] [Code]

Project

THU-BPM/MarkLLM

Public

🏆 First Author & First Contributor
An Open-Source Toolkit for LLM Watermarking (EMNLP 2024 Demo)

Awards

• Outstanding Graduate of Beijing (北京市优秀毕业生)
• Outstanding Graduate of Tsinghua (清华大学优良毕业生)
• Tsinghua Outstanding Graduation Project (清华大学优秀毕业设计)
• Tsinghua Excellent Student Award (清华大学综合优秀奖学金)
• Outstanding Student Cadre of Tsinghua University (清华大学优秀学生干部)

Contact

• Email:
  • panly24@mails.tsinghua.edu.cn
  • panleyi2003@gmail.com

Services

• Program Committee/Reviewer:
  • The Thirteenth International Conference on Learning Representations (ICLR 2025)