I am a PhD student in Computer Science at New York University, advised by Shengjie Wang at the Courant Institute. Before NYU I read Computer Science and Bioinformatics at Shanghai Jiao Tong University, and worked on research at DP Technology, Microsoft Research Asia and the University of Michigan.
My long-term goal is to build AI systems that accelerate the discovery of new therapies and, ultimately, help cure disease and extend human lifespan.
| Direction | What I am after |
|---|---|
| LLM post-training | Reward-matching objectives and replay strategies for GFlowNets and RL fine-tuning — making generative policies cover modes instead of collapsing onto them. |
| Diffusion for drug discovery | Latent and classifier-guided diffusion over 3D molecules: smoother latent spaces, valid geometry, controllable stereochemistry. |
| Foundation models for RNA, proteins & antibodies | Large-scale pretraining that captures evolutionary and structural signal, and benchmarks that test whether representations really encode geometry and chirality. |
Full list on my homepage and Google Scholar.
| Repository | What it is |
|---|---|
| unirna_tf | Uni-RNA — the large-scale pre-trained model for RNA science. |
| ChemGFN | Reward-distribution-matching RL: Rooted Absorbed Prefix Trajectory Balance (ICML 2026). |
- 🧬 PhD Research Intern at ByteDance Seed
- 🎓 PhD in Computer Science, NYU Courant — expected 2030
- 📍 New York, NY
These cards are generated in this repo and refreshed daily by a workflow — no third-party card service involved.