Open to opportunities
M.S. student in Computer Science at the University of Michigan, following a dual-degree B.S. in CS (UMich) and ECE (SJTU). I like building things with AI.
Python
Rust
C/C++
PyTorch
SQL
LaTeX
Latest Posts
Blog
View all →我用 Claude Code 升级 py-alpha-lib：把 101 因子跑到 3.6 秒
这次不是“从零写个 Rust 库”的故事，而是一次更真实的工程升级：我用 Claude Code，把一个能跑的 Python 因子库，升级成了一个能在百万行数据上稳定、快速、可验证的 Py+Ru...
RustPythonQuantPerformanceClaude Code
中文心理问答场景下的 Tokenizer 选型实验
在做中文心理健康问答（PsyQA）相关的研究时，我们发现不同 tokenizer 对下游任务的影响远超预期。这篇文章记录了我们对多种 tokenizer 在中文心理咨询文本上的评测结果。
NLPTokenizerLLM
Spectrum clustering
This is the introduction of spectrum clustering.
clustering
DDPM
This is the introduction of DDPM.
diffusionDDPM
Research
Publications
Google Scholar →Task-Adaptive Tokenization: Enhancing Long-Form Text Generation Efficacy in Mental Health and Beyond
This paper proposed a method to build a better tokenizer for downstream task.
HI-TOM: A Benchmark for Evaluating Higher-Order Theory of Mind Reasoning in Large Language Models
This paper introduce HI-TOM, a Higher Order Theory of Mind benchmark for LLMs.
Builds
Projects
项目页正在整理中,近期会放上几个正在做的东西。