I am a Ph.D. student at University of Chinese Academy of Sciences, dedicated to the field of AI for Science, with a particular emphasis on AI for Drug Discovery (AIDD). I am advised by Professor Xingyu Gao and maintain close collaborations with Professor Peilin Zhao.
From October 2023 to November 2025, I worked as a research intern at Tencent AI Lab, Tencent AI for Life Sciences Lab, and Microsoft Research Asia.
I welcome collaboration opportunities and can be reached at [email protected].
Warning
Problem: The current name of your GitHub Pages repository ("Solution: Please consider renaming the repository to "
http://".
However, if the current repository name is intended, you can ignore this message by removing "{% include widgets/debug_repo_name.html %}" in index.html.
Action required
Problem: The current root path of this site is "baseurl ("_config.yml.
Solution: Please set the
baseurl in _config.yml to "Education
-
University of the Chinese Academy of SciencesElectronic science and technology
Ph.D. StudentMar. 2023 - Jul. 2026 -
University of the Chinese Academy of SciencesElectronic science and technology
M.S. StudentSep. 2020 - Dec. 2022 -
Tianjin UniversityB.S. in Electronic science and technologySep. 2016 - Jul. 2020
Experience
-
Microsoft Research AsiaResearch InternAug. 2025 - Nov. 2025 -
Tencent AI for Life Sciences LabResearch InternJan. 2025 - Jul. 2025 -
Tencent AI LabResearch InternOct. 2023 - Dec. 2024
News
Selected Publications (view all )
A Generative Foundation Model for Antibody Design
Rubo Wang, Fandi Wu, Jiale Shi, Yidong Song, Yu Kong, Jian Ma, Bing He, Qihong Yan, Tianlei Ying, Peilin Zhao, Xingyu Gao, Jianhua Yao
bioRxiv 2025
Further extends IgGM's capabilities to a generative foundation model for antibody design, enabling tasks such as de novo antibody design, affinity maturation, inverse design, structure prediction, and humanization.
A Generative Foundation Model for Antibody Design
Rubo Wang, Fandi Wu, Jiale Shi, Yidong Song, Yu Kong, Jian Ma, Bing He, Qihong Yan, Tianlei Ying, Peilin Zhao, Xingyu Gao, Jianhua Yao
bioRxiv 2025
Further extends IgGM's capabilities to a generative foundation model for antibody design, enabling tasks such as de novo antibody design, affinity maturation, inverse design, structure prediction, and humanization.
Multimodal Mixture-of-Experts with Retrieval Augmentation for Protein Active Site Identification
Jiayang Wu, Jiale Zhou, Rubo Wang#, Xingyi Zhang, Xun Lin, Tianxu Lv, Leong Hou U, Yefeng Zheng# (# corresponding author)
The AAAI Conference on Artificial Intelligence (AAAI) 2026 Oral
MERA is a retrieval-augmented framework for protein active site identification that employs hierarchical multi-expert retrieval and reliability-aware fusion based on Dempster–Shafer theory to overcome data sparsity and modality reliability issues, achieving state-of-the-art performance.
Multimodal Mixture-of-Experts with Retrieval Augmentation for Protein Active Site Identification
Jiayang Wu, Jiale Zhou, Rubo Wang#, Xingyi Zhang, Xun Lin, Tianxu Lv, Leong Hou U, Yefeng Zheng# (# corresponding author)
The AAAI Conference on Artificial Intelligence (AAAI) 2026 Oral
MERA is a retrieval-augmented framework for protein active site identification that employs hierarchical multi-expert retrieval and reliability-aware fusion based on Dempster–Shafer theory to overcome data sparsity and modality reliability issues, achieving state-of-the-art performance.
A synergistic generative-ranking framework for tailored design of therapeutic single-domain antibodies
Yu Kong*, Jiale Shi*, Fandi Wu*, Ting Zhao*, Rubo Wang*, Xiaoyi Zhu, Qingyuan Xu, Yidong Song, Quanxiao Li, Yulu Wang, Xingyu Gao, Yuedong Yang, Yi Feng, Zifei Wang, Weifeng Ge, Yanling Wu, Zhenlin Yang, Jianhua Yao, Tianlei Ying (* equal contribution)
Cell Discovery 2025
Develop TFDesign-sdAb, a synergistic generative-ranking framework integrating the IgGM generative model and A2binder ranking model, which enables single-domain antibodies (sdAbs) to acquire Protein A-binding capability for efficient tag-free purification while preserving their original antigen specificity, with its effectiveness validated by high-resolution structures, providing a generalizable AI-driven solution to advance sdAbs as next-generation biologics.
A synergistic generative-ranking framework for tailored design of therapeutic single-domain antibodies
Yu Kong*, Jiale Shi*, Fandi Wu*, Ting Zhao*, Rubo Wang*, Xiaoyi Zhu, Qingyuan Xu, Yidong Song, Quanxiao Li, Yulu Wang, Xingyu Gao, Yuedong Yang, Yi Feng, Zifei Wang, Weifeng Ge, Yanling Wu, Zhenlin Yang, Jianhua Yao, Tianlei Ying (* equal contribution)
Cell Discovery 2025
Develop TFDesign-sdAb, a synergistic generative-ranking framework integrating the IgGM generative model and A2binder ranking model, which enables single-domain antibodies (sdAbs) to acquire Protein A-binding capability for efficient tag-free purification while preserving their original antigen specificity, with its effectiveness validated by high-resolution structures, providing a generalizable AI-driven solution to advance sdAbs as next-generation biologics.
Geometric Algebra-Enhanced Bayesian Flow Network for RNA Inverse Design
Rubo Wang, Xingyu Gao, Peilin Zhao
Conference on Neural Information Processing Systems (NeurIPS) 2025
Propose RBFN, a geometric-algebra-enhanced Bayesian Flow Network for RNA inverse design that addresses existing method limitations (e.g., narrow structure focus, limited candidates), and it outperforms peers in single/multi-state fixed-backbone benchmarks for effective RNA sequence design.
Geometric Algebra-Enhanced Bayesian Flow Network for RNA Inverse Design
Rubo Wang, Xingyu Gao, Peilin Zhao
Conference on Neural Information Processing Systems (NeurIPS) 2025
Propose RBFN, a geometric-algebra-enhanced Bayesian Flow Network for RNA inverse design that addresses existing method limitations (e.g., narrow structure focus, limited candidates), and it outperforms peers in single/multi-state fixed-backbone benchmarks for effective RNA sequence design.
IgGM: A Generative Model for Functional Antibody and Nanobody Design
Rubo Wang, Fandi Wu, Xingyu Gao, Jiaxiang Wu, Peilin Zhao, Jianhua Yao
International Conference on Learning Representations (ICLR) 2025
IgGM is introduced, a generative model for the de novo design of immunoglobulins with functional specificity that has demonstrated its effectiveness in not only predicting structures but also designing novel antibodies and nanobodies.
IgGM: A Generative Model for Functional Antibody and Nanobody Design
Rubo Wang, Fandi Wu, Xingyu Gao, Jiaxiang Wu, Peilin Zhao, Jianhua Yao
International Conference on Learning Representations (ICLR) 2025
IgGM is introduced, a generative model for the de novo design of immunoglobulins with functional specificity that has demonstrated its effectiveness in not only predicting structures but also designing novel antibodies and nanobodies.