Hi there, I’m Sheldon LI 👋
I am a researcher bridging the gap between Theoretical Computer Science and Generative AI.
- My current research interests focus on Trustworthy AI and Diffusion Models, aiming to make generative models more robust, explainable, and safe.
- Previously, I worked on Problem Reductions and Scientific Computing (specifically Tensor Networks), which gave me a strong foundation in computational complexity and algorithm design.
- Tech Stack:
- AI/ML: PyTorch, Diffusion Models, LLMs
- Scientific Computing: Python，Julia, Typst
- Theory: Computational Complexity
🚀 Selected Projects
- ImageProtection: Integrates as many advarsarial attack algorithm on I2V diffusion model, providing simple and user-friendly interfaces for benchmarking.
- ProblemReductions.jl: A library for computational problem reductions. See our paper: Programming guide for solving constraint satisfaction problems with tensor networks for more details.
