CV

Curriculum Vitae

Research Interests

• Foundation models, generative AI, and multimodal machine learning
• Diffusion models, Transformers, and large-scale model training
• AI for science, biomolecular structure prediction, and protein design

Education

• University of Illinois Urbana-Champaign Urbana, IL
  Ph.D. in Computer Science 2019 – 2024
  Advisor: Jian Peng. Co-advisor: Jianzhu Ma.
• Tsinghua University Beijing, China
  B.E. in Automation 2014 – 2018

Experience

• ByteDance Seed Bellevue, WA
  Senior Research Scientist Jun 2024 – Present
  Core contributor to Protenix, Protenix-v1, and PXDesign.
• University of Illinois Urbana-Champaign Urbana, IL
  Ph.D. Research Assistant Aug 2019 – May 2024
  Developed generative modeling methods for structured 3D molecular data.
• ByteDance AI Lab / AI4Science San Jose, CA
  Research Intern May 2022 – Oct 2022; May 2023 – Aug 2023
  Worked on diffusion-based generative modeling for structure-based drug design.
• Tencent AI Lab Seattle, WA
  NLP Research Intern May 2020 – Aug 2020
  Worked on large-scale language model pretraining and dialogue generation with RL.
• Carnegie Mellon University, Robotics Institute Pittsburgh, PA
  Research Intern Sep 2018 – Jan 2019
  Worked on generative modeling for structured video forecasting and trajectory prediction.

Selected Publications

• Unified Modeling of 3D Molecular Generation via Atomic Interactions with PocketXMol. Cell, 2026.
• LinkerNet: Fragment Poses and Linker Co-Design with 3D Equivariant Diffusion. NeurIPS Spotlight, 2023.
• DecompDiff: Diffusion Models with Decomposed Priors for Structure-Based Drug Design. ICML, 2023.
• 3D Equivariant Diffusion for Target-Aware Molecule Generation and Affinity Prediction. ICLR, 2023.
• Energy-Inspired Molecular Conformation Optimization. ICLR, 2022.
• Generative Hybrid Representations for Activity Forecasting with No-Regret Learning. CVPR Oral, 2020.

For the full list, see Publications.

Technical Skills

• Large-Scale ML: distributed training, mixed-precision, activation checkpointing, memory optimization, custom kernels, inference optimization
• Modeling: foundation models, Transformers, diffusion models, multimodal modeling, structured generation, geometric deep learning, reinforcement learning
• Programming: Python, PyTorch, C/C++, Linux, NumPy/SciPy