Zequn Li 李泽群

Incoming Ph.D. student · Tsinghua University

I am an incoming direct-entry Ph.D. student in the Department of Computer Science and Technology at Tsinghua University (expected start: 2027). My advisor is Prof. Jiannan Wang. My undergraduate major is Computer Science and Technology (Elite Class) at Beijing Institute of Technology.

My research interests lie at the intersection of data systems and large language models, with related work in natural language processing and computer vision. I enjoy building efficient, practical AI systems.

中文简介

我是李泽群,已获清华大学计算机科学与技术系直博录取,预计于 2027 年入学,导师为王健楠老师。本科就读于北京理工大学计算机科学与技术拔尖班。我的研究兴趣主要包括数据系统与大语言模型。目前,我正在探索异构架构上的高效多模态数据处理,重点关注 Apple Silicon 的统一内存架构、CPU–GPU 协同、资源竞争与自适应执行策略。

Zequn Li wearing an orange shirt outdoors
Beijing, China

Current Research

Multimodal Data Processing on Heterogeneous Systems

I am currently exploring efficient execution of multimodal data processing pipelines on heterogeneous architectures, with a particular focus on Apple Silicon's unified memory architecture. My research investigates CPU-GPU collaboration, resource contention, and adaptive execution strategies to improve end-to-end throughput.

News

More updates
  • My undergraduate project on efficient instruction-tuning data extraction received support from the Beijing Natural Science Foundation.

All Research

Google Scholar

DocDB framework showing two-level index construction, evidence-augmented segment retrieval, and query optimization

DocDB: A Database for Unstructured Document Analysis

Zequn Li, Yuanhao Zhong, Chengliang Chai, Zhaoze Sun, Yuhao Deng, Ye Yuan, Guoren Wang, Lei Cao

PVLDB 2025Demo

Query unstructured documents with SQL-like operations. Two-level retrieval and document-specific execution plans reduce the amount of text and the number of calls sent to LLMs.

Qiyan framework with document clustering, adaptive sampling, QA extraction, contrastive learning, and optimal-transport evaluation

Efficient Instruction-Tuning Data Extraction for Large Language Models

Project lead Qiyan Program (启研计划)

Extract high-value QA pairs for LLM instruction tuning without scanning every document. Contrastive representations, adaptive sampling, and optimal-transport feedback guide where to extract next.

Beijing Natural Science Foundation · CNY 50,000 research grant

Education

Honors & Awards

Academic honors

  • National Scholarship Top 1%
  • Huawei Scholarship Top 1%
  • BIT First-class Scholarship 5 times
  • BIT Outstanding Student 2 times

Selected competitions

  • Mathematical Contest in Modeling Meritorious Winner
  • CCL Factivity Inference Evaluation National 1st Place
  • Chinese Mathematics Competitions National 1st Prize
  • BIT “Century Cup” Competition Grand Prize
  • CUMCM Provincial 2nd Prize
  • National English Competition for College Students National 3rd Prize