CV
� View PDF version
Profile
- First-year undergraduate in School of Mathematical Sciences at Fudan University, pursuing a double degree in Information and Computing Science & Artificial Intelligence.
- GPA: 3.96/4.00 (most recent semester); ranked 10th (top 5%) in the cohort.
- Interested in the mathematical foundations of AI, large language models, reinforcement learning, algorithms, and efficient model training.
- Most listed projects were completed independently outside a formal lab or research group.
Education
- Fudan University, Shanghai, China
- Double Degree in Information and Computing Science & Artificial Intelligence
- GPA: 3.96/4.00 (most recent semester)
- Rank: 10th (top 5%) in the cohort.
Research Interests
- Algorithms and optimization
- Reinforcement Learning
- Deep Learning
- LLM reasoning
Selected Open-Source Projects
- TinyLoRA-GRPO-Coder
- Independent open-source reimplementation and adaptation of TinyLoRA + GRPO from Learning to Reason in 13 Parameters
- Migrated the method from math reasoning to verifiable competitive-programming code generation on Qwen2.5-Coder-3B
- Trained with only a tiny number of shared trainable parameters and real compile-and-run rewards instead of static heuristics
- Built the full pipeline end to end: data processing, training, multi-GPU setup, reward design, evaluation, and validation
- NAD Next
- Collaborative framework for analyzing large-language-model neuron activations and reasoning processes
- Covers activation-cache construction, selector evaluation, token-level statistics, and visualization
- My primary contribution focuses on algorithm construction and method design
- Work in progress: due to practical constraints, some modules are not yet publicly available and will be released progressively
Skills
- Programming: Python, C++, C, Java, Node.js, Lean 4
- ML/AI: PyTorch, deep learning, LLM fine-tuning, reinforcement learning, computer vision
- Tools: Git, GitHub, Linux, VS Code, Markdown, LaTeX
- Other strengths: self-directed learning, mathematical problem solving, technical writing, competitive programming
