Xiao
Hong
RenShixin
Ren Shixin
任世鑫
Tsinghua University, Zhili College, Information and Computational Science | Undergraduate
Hi! I am an undergraduate student (Class of 2023) majoring in Information and Computational Science at Zhili College, Tsinghua University. I am currently doing research at the Natural Language Processing Lab (THUNLP), Tsinghua University. My research interests include LLM Algorithms, LLM Systems, and High-Performance Computing (HPC).
Education
Tsinghua University · Zhili College
2023.09 -- Present
Nankai High School, Tianjin
2020.09 -- 2023.08
Projects
Chinese Pre-training Data Synthesis via Web Rewriting
2025.10 -- Present
Synthesized high-quality Chinese data using Ultra-FineWeb-zh as source data, following the Nemotron-CC methodology. Continual pre-training of a 100B-token MiniCPM model on 3B synthetic tokens showed significant improvements on MMLU, CMMLU, and C-Eval benchmarks.
MiniCPM-SALA
2026.01 -- 2026.02
Prepared general knowledge data for the fine-tuning stage of MiniCPM-SALA.
Experience
Tsinghua University HPC Team
2025.06 -- Present
Tsinghua University Algorithm Association
2023.09 -- Present
NOI 2026 Winter Camp
2026.02
Object-Oriented Programming
Spring 2024
Zhili-Math&CS Class 31
2023.09 -- Present
Zhili College Student Union
2023.09 -- 2024.08
Competitions & Awards
- PAC2025 (National Parallel Application Challenge): 1st Place
- Excellence Scholarship for Sci-Tech Innovation: 2024, 2025
- 2023 CCPC (Collegiate Programming Contest, Harbin): 4th Place
Skills
- Programming Languages: C++, Python, Rust
- Domains: LLM, HPC, Network
- Current Interests: LLM Algorithms, LLM Systems
Blog: Posts · Tags · Categories · Archive