About Me
M.S. in Electrical Engineering @ Stanford University
Software Engineering · Machine Learning · Artificial Intelligence
I’m an Electrical Engineering master’s student at Stanford University with a background in software development, machine learning, and computer vision. My work spans LLM interpretability, reusable image-restoration code, and an iOS app for recording and reflecting on everyday life. I’m interested in building AI-powered products and applying machine learning to real-world problems.
📌 Seeking Summer 2027 Software Engineering & AI/ML Engineering Internships.
[GitHub] · [LinkedIn] · [Email] · [Google Scholar] · [Calendar Demo] · [LLM Interpretability Blog]
📖 Education
- Stanford University — Master of Science in Electrical Engineering, Sep 2026 – Jun 2028 (Expected).
- Nanjing University — Bachelor of Science in Intelligence Science and Technology, Sep 2022 – Jun 2026. GPA: 4.53/5.00 (Top 3%).
💼 Experience
Interpretability of Large Language Models
Research Assistant — Yale University · Jul 2025 – Sep 2026
Developed an end-to-end LLM interpretability pipeline integrating cross-layer transcoder features, attribution analysis, and circuit tracing. Designed sparse feature representations achieving 40× higher sparsity while preserving model performance, and conducted controlled intervention experiments to analyze model behavior.
Blog: What Makes a Language Model Keep Reasoning · [source]
Image Restoration and Inverse Problems
Researcher — Nanjing University · Oct 2024 – Aug 2025
Developed the official open-source implementation of Reverse Convolution, which has received over 100 GitHub stars. Implemented a reusable closed-form reverse-convolution operator and a modular evaluation pipeline for image restoration and deblurring.
Publication: Reverse Convolution and Its Applications to Image Restoration — ICCV 2025, Co-first Author.
Computer Vision and Facial Privacy Protection
Undergraduate Researcher — Nanjing University · Aug 2024 – Apr 2025
Developed a hierarchical adversarial optimization pipeline to generate protective perturbations against customized text-to-image synthesis, including reward-guided optimization and evaluation workflows.
Preprint: Anti-Aesthetics: Protecting Facial Privacy against Customized Text-to-Image Synthesis — arXiv, 2025 (withdrawn).
💻 Selected Projects
Done — AI-Powered Time Tracking & Reflection iOS App
Independently developed an iOS app that brings timelines, interruption tracking, and activity reflection into one place. Its LLM assistant uses structured tool calling to query schedules and manage events and to-dos, with persistent conversation history and user confirmation for deletions. Implemented local JSON persistence, Supabase cloud synchronization, backup recovery, user-reviewed conflict resolution, and unit tests for timeline interaction logic.
Swift · SwiftUI · UIKit · Supabase · LLM Integration · [calendar demo]
📝 Publications
Anti-Aesthetics: Protecting Facial Privacy against Customized Text-to-Image Synthesis
Songping Wang, Yueming Lyu, Shiqi Liu, Ning Li, Tong Tong, Hao Sun, Caifeng Shan
Withdrawn preprint · [arXiv record]
🛠 Technical Skills
- Programming Languages: Python, Swift, SQL, C/C++
- Machine Learning & AI: PyTorch, NumPy, Hugging Face Transformers, Large Language Models, Computer Vision, Model Evaluation
- Software Development: SwiftUI, UIKit, Git, Linux, Supabase, Unit Testing
🎖 Honors and Awards
- 2026, Nanjing University Outstanding Graduate.
- 2024, National First Prize, 26th China Robotics and Artificial Intelligence Competition.
- 2023, Guo Xie Biyong Scholarship (Top Prize).