Do interesting things

中文

Basic Info


I am currently an undergraduate student at the School of Mathematical Sciences, Fudan University, pursuing a double degree in Information and Computing Science and Artificial Intelligence.
At the same time, I have been actively self-studying interesting topics across computer science.
I am exploring the intersection I truly want to work on and study.
I hope to contribute to making society better.

My repos

  • The following projects were primarily completed independently, with AI used only as an auxiliary tool where appropriate. These works were not conducted under a laboratory or research group; rather, they reflect my self-directed exploration, sustained learning, and independent implementation outside a formal research environment.
    1. TinyLoRA-GRPO-Coder
      An independent open-source reimplementation and adaptation of TinyLoRA + GRPO from Learning to Reason in 13 Parameters, migrated from math reasoning to verifiable competitive-programming code generation. Built on Qwen2.5-Coder-3B with only a tiny number of shared trainable parameters, the project uses real compile-and-run rewards rather than static heuristics. I developed the full pipeline end to end, including data processing, training, multi-GPU setup, reward design, evaluation, and validation, which significantly strengthened my ability to turn a paper into a working research system.

    2. KaggleCompetitions
      Participated in several Kaggle competitions (see repo: KaggleCompetitions), gaining broad exposure to and practical experience with various machine learning tools.

    3. Hone My C Plus Plus
      Explorations of Advanced Algorithms and Modern C++ can be found in Hone My C Plus Plus.

    4. microgpt.cpp
      A simple microgpt.cpp in 300 lines (see repo: microgpt.cpp).

    5. Baseball
      A Strike/Ball Classification Model using CNN + ResNet18 for baseball pitch analysis (see repo: Baseball).

    6. Sample Java
      The essence of Randomized Algorithms lies in the “Art of Sampling”—making choices that appear random but are statistically effective. This repo also marks my first taste of Java, hence the name Sample Java. The majority of the repo is my study notes.
    7. DeepLearning, GenerativeModel, ReinforcementLearning
      These three repositories serve as my learning records, covering the journey from theoretical derivation to code implementation in Deep Learning, Generative Models, and Reinforcement Learning (see repos: DeepLearning, GenerativeModel, ReinforcementLearning).
  • The following project was completed in collaboration with others.
    1. NAD Next A collaborative framework for analyzing large-language-model neuron activations and reasoning processes. The project covers activation-cache construction, selector evaluation, token-level statistics, and visualization. Our goal is to compare different runs on the same problem via CoT-, activation-, and ensemble-based signals, and estimate which run is more likely to be correct or incorrect. My primary contribution to this project is algorithm construction and method design.

      Note: This project is currently a work in progress (WIP). Due to practical constraints, some content cannot be open-sourced on GitHub immediately; therefore, the current public repository is not yet complete. We will continue to add materials and update toward a more complete release as conditions permit.


Service and Community Involvement

Beyond my personal projects, I also contribute to community-oriented open-source work.

  1. FDUGuideBook/nav-site
    Contribute to this navigation site for the Fudan community continuously.

Tech stack and tools

DomainSkills
Language Python   Node.js   C++   C   Java   Lean 4   ε-N language, ε-δ language
IDE VS Code   📕Draftbooks
OS Windows   Linux
Other Markdown   LaTeX   Redstone

Education

School of Mathematical Sciences, Fudan University

When clouds gather, the mountain grows lovelier still; when they part, it stands like a painting.
Clouds lend it shadow and light, and give shape to its height.