I am Lee Han Jin, a researcher in artificial intelligence security and generative model reliability, currently pursuing an integrated M.S./Ph.D. in Computer Science at Yonsei University. My research focuses on understanding, controlling, and securing modern generative models, particularly diffusion-based systems, from a principled and mathematical perspective.

At AISLab, I work on advancing AI pathway security, adversarial robustness in generative models, and distribution-preserving control mechanisms. My work aims to bridge theoretical rigor and practical system reliability, ensuring that AI systems behave safely and predictably even under distribution shifts or adversarial conditions.

Beyond model design, I am deeply interested in AI system infrastructure, research tooling, and scalable evaluation frameworks that improve how research labs operate and validate AI technologies. I approach problems with a strong emphasis on structure, first-principles reasoning, and long-term impact.

My goal is to build AI systems that are not only powerful, but trustworthy, controllable, and secure by design.