Jiahan Li / Ced

Hi, I am Jiahan Li (李嘉涵), also known as Ced—a name inspired by the CED family of proteins. I am a CS Ph.D. student at NYU Courant, advised by Rob Fergus and Kyunghyun Cho.

Previously, I earned my B.S. degree from Yuanpei College at Peking University in 2023. I also spent three years at Helixon (Earendil Labs), where I worked with Jian Peng and Jianzhu Ma on full-stack high-throughput antibody design and screening. In addition, I work closely with Wenhui Duan on AI for materials discovery. I studied biology in high school and won a gold medal at the Chinese National Biology Olympiad.

Research Interests
I build generative models and agentic systems for multimodal intelligence and AI for Science, with a focus on:

• Unified Multimodal Foundation Models: Native pre-training and post-training paradigms unifying perception and generation across language, vision, and molecular structures.
• Multimodal Agentic Systems: General-purpose agents capable of cross-modal reasoning (vision, DNA/genomics) and complex task generation (visual synthesis, drug discovery).
• Lab-in-the-Loop Scientific Discovery: LLMs and deep generative models for de novo biomolecular (proteins, RNA) and chemical (small molecules, materials) design integrated with wet-lab validation.
• Biosecurity & AI Alignment: Safety evaluation, risk mitigation, and governance for frontier AI systems in biology and chemistry.

Outside research, I'm a Hip-Hop producer and DJ. Some prod by me: PKU Cypher 2025, PKU Cypher 2023.