CV
Research Profile
Machine learning researcher focused on biomolecular foundation models, structure, gene expression, and multimodal biology. Experience in self-supervised learning and distributed training with PyTorch, Megatron-LM, and Hugging Face, including pretraining on 64 NVIDIA A100 GPUs.
- Core contributor to 1.6B-parameter GB.RNA and 16B-parameter Mixture-of-Experts GB.Protein.
- Stanford RNA 3D Folding (Kaggle, 2025): Gold Medal, 6th/1,516 teams.
Education
PhD in Machine Learning — Mohamed bin Zayed University of Artificial Intelligence
Abu Dhabi, UAE · 2022–Dec 2026 (expected)
- GPA: 3.95/4.00; full sponsorship.
- Thesis: Representation Learning with Foundation Models for Biomolecular Structure and Function Prediction (in progress).
- Advisor: Prof. Le Song.
- Thesis committee: Eric P. Xing, Kun Zhang, and Christoph Feinauer.
MSc in Computer Science — University of Chinese Academy of Sciences
Beijing, China · 2019–2022
- GPA: 3.86/4.00; academic scholarship.
- Thesis: Research on Decoding Natural Language from Functional Brain Images.
- Advisor: Prof. Jiajun Zhang.
BSc in Statistics — Sun Yat-sen University
Guangzhou, China · 2010–2014
- GPA: 4.0/5.0; top 10%.
Research Experience
Research Intern — GenBio AI
Abu Dhabi, UAE · Dec 2024–Dec 2025; Jun 2026–present (expected end: Dec 2026)
- Developed GB.RNA and GB.RNA-CDS for noncoding RNA and coding sequences; evaluated them on 100+ structure, expression, and function tasks.
- Benchmarked GB.Protein, GB.Protein-RAG, and GB.Protein2StructureToken on ProteinGym DMS and xTrimoPGLM for protein understanding and fitness prediction.
- Curated a GTEx transcript isoform dataset; built multimodal models using DNA, RNA, and protein data to predict tissue-specific isoform expression.
- Added sequence-level tasks and multimodal fusion to GB.ModelGenerator.
- Collaborated on wet-lab validation of designed RNA aptamers with the RNA design team and an external CRO.
Visiting PhD Student — Sailing Lab, Carnegie Mellon University
Pittsburgh, USA · Jun–Sep 2024 · Host: Prof. Eric Xing
- Studied pretraining data and strategies for billion-parameter models spanning noncoding and protein-coding RNA.
- Explored efficient pretraining and evaluation of a 16B-parameter protein language model.
Algorithm Intern — BioMap
Beijing, China · Jul–Sep 2023
- Pretrained cross-modal protein models on STRING interaction data with Megatron-LM for antibody–antigen binding affinity prediction.
Research Assistant — Institute of Automation, Chinese Academy of Sciences
Beijing, China · Jul 2019–May 2022
- Developed fMRI-to-word and fMRI-to-text decoding tasks and methods using pretrained encoder–decoder models.
Selected Publications and Preprints
- Shuxian Zou, Tianhua Tao, Sazan Mahbub, Caleb Ellington, Robin Algayres, Dian Li, Yonghao Zhuang, Hongyi Wang, Le Song, and Eric P. Xing. A large-scale foundation model for RNA function and structure prediction. Under review at Nature Methods. NeurIPS 2024 AIDrugX Workshop (Spotlight). [Paper] [GitHub] [Hugging Face]
- Ning Sun, Shuxian Zou, Tianhua Tao, Sazan Mahbub, Dian Li, Yonghao Zhuang, Hongyi Wang, Xingyi Cheng, Le Song, and Eric P. Xing. Mixture of experts enable efficient and effective protein understanding and design. Under review at Nature Methods. NeurIPS 2024 AIDrugX Workshop (Spotlight). [Paper] [GitHub] [Hugging Face]
- Shuxian Zou, Jiayou Zhang, Bingkang Zhao, Hui Li, Eric P. Xing, and Le Song. Improving RNA 3D Structure Prediction via Language Model-Augmented AlphaFold 3. NeurIPS 2025 MLSB Workshop. [Paper]
- Youhan Lee, Shujun He, Toshiyuki Oda, …, Shuxian Zou, …, David Baker, …, Rhiju Das. Template-based RNA structure prediction advanced through a blind code competition. Accepted in principle at Nature Methods. [Paper]
- Caleb Ellington, Dian Li, Shuxian Zou, Elijah Cole, Ning Sun, Sohan Addagudi, Le Song, and Eric P. Xing. Rapid and Reproducible Multimodal Biological Foundation Model Development with AIDO.ModelGenerator. ICML 2025 GenBio Workshop (Spotlight). [Paper] [GitHub]
- Shuxian Zou, Hui Li, Shentong Mo, Xingyi Cheng, Eric P. Xing, and Le Song. Linker-Tuning: Optimizing Continuous Prompts for Heterodimeric Protein Prediction. arXiv preprint, 2023. [Paper]
- Shuxian Zou, Shaonan Wang, Jiajun Zhang, and Chengqing Zong. Cross-Modal Cloze Task: A New Task to Brain-to-Word Decoding. Findings of ACL 2022, pp. 648–657. [Paper] [GitHub]
- Shuxian Zou, Shaonan Wang, Jiajun Zhang, and Chengqing Zong. Towards Brain-to-Text Generation: Neural Decoding with Pre-trained Encoder-Decoder Models. NeurIPS 2021 AI for Science Workshop, Dec 13, 2021. [Paper]
Teaching and Additional Experience
- Teaching Assistant, MBZUAI:
- Advanced Machine Learning (Spring 2025) — Instructors: Salem Lahlou, Tongliang Liu, Michalis Vazirgiannis, and Yuanzhi Li
- Machine Learning with Python (Fall 2025, 2026) — Instructor: Maxim Panov
- Data Analyst, China Mobile, Guangzhou, China — 2014–2018.
- Customer analysis and KPI project management.
Selected Awards
- Gold Medal, Stanford RNA 3D Folding (Kaggle, 2025) — 6th/1,516 teams. Solution.
- Outstanding Teaching Assistant Award, MBZUAI, Spring 2025.
Skills
- Programming: Python, Shell, SQL.
- Machine learning: PyTorch, Megatron-LM, Hugging Face, Lightning, TensorFlow, scikit-learn.
- HPC and infrastructure: SLURM, Weights & Biases (W&B), distributed multi-GPU training.
- Developer tools: Git, Claude Code, Codex.
- Languages: English (professional proficiency), Chinese (native), Cantonese (fluent), Hakka (native).