CV
Education
- Master of Engineering in Electronic Information (Computer Science and Technology), Tsinghua University, 2028 (expected)
- Advisor: Associate Professor Jianfei Chen
- Research Lab: TSAIL
- Bachelor of Engineering in Computer Science and Technology, Tsinghua University, 2026 (expected)
- Major GPA: 3.76/4.0
- Relevant Coursework (Grade: A): Ordinary Differential Equations; Probability and Statistics; Fundamentals of Programming; Programing and Training; Software Engineering; Operating Systems; Principles of Signal Processing; Database Special Topic Training
- Previous Major: Medicine. Pivoted to Computer Science to pursue a passion for Artificial Intelligence.
Research Statement
- My research focuses on improving the long context capability of large language models (LLMs).
Publications
- Scaling External Knowledge Input Beyond Context Windows of LLMs via Multi-Agent Collaboration
Zijun Liu*, Zhennan Wan*, Peng Li, Ming Yan, Ji Zhang, Fei Huang, Yang Liu
arXiv 2025. Under review. [Paper] [GitHub] [Data]
*Equal contribution.
Research Experience
- September 2024 - May 2025: Natural Language Processing Lab, Tsinghua University (THUNLP)
- Advisor: Associate Professor Peng Li and Professor Yang Liu
- Research Topic: Scaling External Knowledge Input Beyond the Context Length of LLMs via Multi-Agent Collaboration
- Role: Co-leader
- Description: Developed a multi-agent framework, ExtAgents, to overcome the context window limitations of current large language models (LLMs) and enable better scalability in inference-time knowledge integration without longer-context training.
- February 2024 - August 2024: 3D Visual Computing and Machine Intelligence (3DVICI) Lab, Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University
- Advisor: Assistant Professor Li Yi
- Research Topic: Editing Human Videos for Robotic Skill Training
- Role: Co-leader
- Description: Utilized video editing techniques to convert human hand-object interaction videos into robotic hand-object interaction videos; extracted 6D poses of human videos; compared human data with robot data, referencing pipelines such as OpenVLA, to validate the effectiveness of human videos for robotic training.
- July 2023 - August 2023: State Key Laboratory of Intelligent Technology and Systems, Department of Computer Science and Technology, Tsinghua University
- Advisor: Assistant Professor Bin Fang
- Research Topic: Human-Machine Collaborative Operation of Super-functional Prosthetic Hands
- Role: Project Member
- Description: Trained neural network models to extract human hand movements from electromyographic signals of the human arm.
Internship Experience
- June 2025 - August 2025: Code Intelligence Center, Technology and Engineering Group (TEG), Tencent
- Project Topic: Applying CodeLLMs for Multi-Point Code Completion and Smart Rewriting
- Role: Student Leader
- Description: Researched data synthesis algorithms for CodeLLMs on cursor prediction and intelligent rewrite tasks; trained a CodeLLM-based fusion model to implement cursor prediction and intelligent rewrite suggestions during code editing. Our group was ranked 1st among 5 groups.
Course Project
- April 2025 - June 2025: Frontiers in AI safety and Governance (2025 Spring)
- Lecturer: Professor Peng Cui
- Project Topic: Adaptive Safety Priming: Inference-Time Safeguards for Large Reasoning Models
- Role: Independent
- Description: Developed a lightweight and dynamic safety mechanism, Adaptive Safety Priming (ASP), for large reasoning models (LRM) at inference time, which leverages their step-by-step inference process to enable real-time intervention. This approach provides a more adaptive and resource efficient path to develop robustly safe models. Report
Skills
- Programming
- C/C++, Python, Pytorch
- Tools
- Git, Linux, Django, Docker, LaTeX
- Language
- College English Test-6 (CET6): 601 / 710 (top 10% of the normative group)
Awards
- Tsinghua University Software Engineering Outstanding Project Award (2024)
