Guolin Ke

About Me

Guolin Ke is a machine learning researcher. He previously served as Senior Vice President of AI at DP Technology (2021–2026) and as a Senior Researcher in the Machine Learning Group at Microsoft Research Asia (MSRA) (2016–2021).

His research spans high-performance machine learning algorithms and systems, large-scale pretrained and generative models, and AI for Science. His work includes molecular representation learning, 3D geometric modeling and generation, protein structure prediction, and scientific document understanding. His publications have received over 42,000 citations on Google Scholar, and his open-source projects have been widely adopted across academia and industry.

In 2016, he created LightGBM, a widely used gradient-boosted decision tree (GBDT) framework, during his internship at MSRA. He later led the development of Graphormer for graph representation learning and Uni-Fold, an open-source platform for protein structure prediction with publicly available training code and datasets.

His work in molecular modeling includes Uni-Mol, a framework for 3D molecular representation learning; Uni-Mol+, which combines 3D conformation refinement with quantum chemical property prediction; and Uni-Mol2, which explores molecular pretraining at scale. His work also includes Uni-3DAR, a unified autoregressive framework for 3D generation and understanding, and MolParser, an end-to-end system for recognizing chemical structures, including Markush structures, from scientific literature and patents.

His competition achievements include first place in the Alibaba (Tmall) Big Data Competition (2014; RMB 1 million prize) and the Didi Algorithm Competition (2016; US$100,000 prize). The Graphormer and Graphormer-3D projects he led also won first place in the quantum prediction track of the OGB Large-Scale Challenge at KDD Cup 2021 and the Open Catalyst Challenge at NeurIPS 2021, respectively.

News

• Jan 22, 2026, our paper Hyperspherical Latents Improve Continuous-Token Autoregressive Generation was accepted at ICLR 2026.

• Dec 17, 2025, we released the Uni-Parser Technical Report, describing a document parsing engine for scientific literature and patents.

• Mar 21, 2025, we released Uni-3DAR, an autoregressive model that unifies a wide range of 3D tasks, from micro-scale to macro-scale structures.

• Nov 11, 2023, LightGBM was selected in AI100: Top 100 AI Achievements (1943-2021).

• Oct 4, 2023, Guolin was selected among the World’s Top 2% Scientists.

• Jul 7, 2023, Uni-Mol+ won first place in the Open Catalyst 2020 IS2RE Direct benchmark.

• Mar 16, 2023, Uni-Mol+ won first place in the OGB Large-Scale Challenge PCQM4Mv2.

• Feb 16, 2023, we released Uni-Fold MuSSe, a de novo protein complex prediction tool that requires only a single sequence input.

• Aug 31, 2022, we released Uni-Fold Symmetry, supporting the end-to-end prediction of extremely large protein complexes.

• Aug 6, 2022, we released the PyTorch version of Uni-Fold, an AlphaFold and AlphaFold-Multimer reimplementation with open-source training code and datasets.

• Jul 21, 2022, LightGBM became faster through quantized training and a new GPU version. See our paper for more information.

• Jun 10, 2022, we released Uni-Mol, the first 3D molecular/pocket pretraining framework, outperforming prior state-of-the-art methods in molecular property prediction, protein-ligand binding pose prediction, and molecular conformation generation.

• Dec 8, 2021, Graphormer-3D won first place in the Open Catalyst Challenge at NeurIPS 2021.

• Jun 17, 2021, Graphormer won first place in the OGB Large-Scale Challenge PCQM4M-LSC at KDD Cup 2021.

Publications

Selected publications are listed below; the full list is available at Google Scholar.

• Guolin Ke, Hui Xue. “Hyperspherical Latents Improve Continuous-Token Autoregressive Generation”. ICLR 2026.

• Yongqi Jin, Jun-Jie Wang, Fanjie Xu, Xiaohong Ji, Zhifeng Gao, Linfeng Zhang, Guolin Ke, Rong Zhu, Weinan E. “NMR-Solver: Automated Structure Elucidation via Large-Scale Spectral Matching and Physics-Guided Fragment Optimization”. Nature Communications 2026.

• Xi Fang, Haocheng Lu, Han Lyu, Chengxiang Luo, Linfeng Zhang, Guolin Ke. “MolParser-Mobile: Ultrafast OCSR System for Large-Scale Chemical Literature Mining”. arXiv preprint arXiv:2609.05807, 2026.

• Jingjie Ning, Xiaochuan Li, Ji Zeng, Chenyan Xiong, Guolin Ke. “Closed-loop Auto Research for Molecular Property Prediction: Discovering and Certifying Generalizable Improvements”. arXiv preprint arXiv:2606.22731, 2026.

• Yongqi Jin, Yecheng Wang, Jun-jie Wang, Rong Zhu, Guolin Ke, Weinan E. “From Human Labels to Literature: Semi-Supervised Learning of NMR Chemical Shifts at Scale”. ICML 2026.

• Shuqi Lu, Xiaohong Ji, Bohang Zhang, Lin Yao, Siyuan Liu, Zhifeng Gao, Linfeng Zhang, Guolin Ke. “Beyond Atoms: Enhancing Molecular Pretrained Representations with 3D Space Modeling”. ICML 2025.

• Sihang Li, Jin Huang, Jiaxi Zhuang, Yaorui Shi, Xiaochen Cai, Mingjun Xu, Xiang Wang, Linfeng Zhang, Guolin Ke, Hengxing Cai. “SciLitLLM: How to Adapt LLMs for Scientific Literature Understanding”. ICLR 2025.

• Xi Fang, Jiankun Wang, Xiaochen Cai, Shangqian Chen, Shuwen Yang, Haoyi Tao, Nan Wang, Lin Yao, Linfeng Zhang, Guolin Ke. “MolParser: End-to-end Visual Recognition of Molecule Structures in the Wild”. ICCV 2025.

• Qingsi Lai, Fanjie Xu, Lin Yao, Zhifeng Gao, Siyuan Liu, Hongshuai Wang, Shuqi Lu, Di He, Liwei Wang, Linfeng Zhang, Cheng Wang, Guolin Ke. “End-to-End Crystal Structure Prediction from Powder X-Ray Diffraction”. Advanced Science 2025.

• Xi Fang, Haoyi Tao, Shuwen Yang, Chaozheng Huang, Suyang Zhong, Haocheng Lu, Han Lyu, Junjie Wang, Xinyu Li, Linfeng Zhang, Guolin Ke. “Uni-Parser Technical Report”. arXiv preprint arXiv:2512.15098, 2025.

• Zhen Wang, Zhifeng Gao, Guolin Ke. “Masked-and-Reordered Self-Supervision for Reinforcement Learning from Verifiable Rewards”. arXiv preprint arXiv:2511.17473, 2025.

• Haoyu Dong, Pengkun Zhang, Mingzhe Lu, Yanzhen Shen, Guolin Ke. “MachineLearningLM: Scaling Many-shot In-context Learning via Continued Pretraining”. arXiv preprint arXiv:2509.06806, 2025.

• Shuqi Lu, Haowei Lin, Lin Yao, Zhifeng Gao, Xiaohong Ji, Yitao Liang, Weinan E, Linfeng Zhang, Guolin Ke. “Unified Cross-Scale 3D Generation and Understanding via Autoregressive Modeling”. arXiv preprint arXiv:2503.16278, 2025. (Uni-3DAR)

• Hengxing Cai, Xiaochen Cai, Junhan Chang, Sihang Li, Lin Yao, Wang Changxin, Zhifeng Gao, Hongshuai Wang, Li Yongge, Mujie Lin, Shuwen Yang, Jiankun Wang, Mingjun Xu, Jin Huang, Xi Fang, Jiaxi Zhuang, Yuqi Yin, Yaqi Li, Changhong Chen, Zheng Cheng, Zifeng Zhao, Linfeng Zhang, Guolin Ke. “SciAssess: Benchmarking LLM Proficiency in Scientific Literature Analysis”. Findings of NAACL 2025.

• Eric Alcaide, Zhifeng Gao, Guolin Ke, Yaqi Li, Linfeng Zhang, Hang Zheng, Gengmo Zhou. “Uni-Mol Docking V2: Towards Realistic and Accurate Binding Pose Prediction”. ICANN 2025, AI in Drug Discovery (AIDD) Workshop, 34–41.

• Gengmo Zhou, Zhen Wang, Feng Yu, Guolin Ke, Zhewei Wei, Zhifeng Gao. “S-MolSearch: 3D Semi-supervised Contrastive Learning for Bioactive Molecule Search”. NeurIPS 2024.

• Xiaohong Ji, Zhen Wang, Zhifeng Gao, Hang Zheng, Linfeng Zhang, Guolin Ke, Weinan E. “Exploring Molecular Pretraining Model at Scale”. NeurIPS 2024. (Uni-Mol2)

• Shuqi Lu, Zhifeng Gao, Di He, Linfeng Zhang, and Guolin Ke. “Data-driven Quantum Chemical Property Prediction Leveraging 3D Conformations with Uni-Mol+”. Nature Communications 2024.

• Lin Yao, Wentao Guo, Zhen Wang, Shang Xiang, Wentan Liu, Guolin Ke. “Node-Aligned Graph-to-Graph: Elevating Template-free Deep Learning Approaches in Single-Step Retrosynthesis”. JACS Au 2024, 4, 3, 992-1003.

• Shuqi Lu, Lin Yao, Xi Chen, Hang Zheng, Di He, and Guolin Ke. “3D Molecular Generation via Virtual Dynamics”. Transactions on Machine Learning Research (TMLR) 2024.

• Jinhua Zhu, Zhenyu He, Ziyao Li, Guolin Ke, and Linfeng Zhang. “Uni-Fold MuSSe: De Novo Protein Complex Prediction with Protein Language Models”. bioRxiv 2023.

• Yuejiang Yu, Shuqi Lu, Zhifeng Gao, Hang Zheng, and Guolin Ke. “Do Deep Learning Models Really Outperform Traditional Approaches in Molecular Docking?”. arXiv preprint arXiv:2302.07134, 2023.

• Gengmo Zhou, Zhifeng Gao, Zhewei Wei, Hang Zheng, and Guolin Ke. “Do Deep Learning Methods Really Perform Better in Molecular Conformation Generation?”. arXiv preprint arXiv:2302.07061, 2023.

• Lin Yao, Ruihan Xu, Zhifeng Gao, Guolin Ke, and Yuhang Wang. “Boosted Ab Initio Cryo-EM 3D Reconstruction with ACE-EM”. arXiv preprint arXiv:2302.06091, 2023.

• Gengmo Zhou, Zhifeng Gao, Qiankun Ding, Hang Zheng, Hongteng Xu, Zhewei Wei, Linfeng Zhang, and Guolin Ke. “Uni-Mol: A Universal 3D Molecular Representation Learning Framework”. ICLR 2023.

• Ziyao Li, Shuwen Yang, Xuyang Liu, Weijie Chen, Han Wen, Fan Shen, Guolin Ke, and Linfeng Zhang. “Uni-Fold Symmetry: harnessing symmetry in folding large protein complexes”. bioRxiv 2022.

• Ziyao Li, Xuyang Liu, Weijie Chen, Fan Shen, Hangrui Bi, Guolin Ke, and Linfeng Zhang. “Uni-Fold: An Open-Source Platform for Developing Protein Folding Models beyond AlphaFold”. bioRxiv 2022.

• Yu Shi, Guolin Ke, Zhuoming Chen, Shuxin Zheng, and Tie-Yan Liu. “Quantized Training of Gradient Boosting Decision Trees”. NeurIPS 2022.

• Yu Shi, Shuxin Zheng, Guolin Ke, Yifei Shen, Jiacheng You, Jiyan He, Shengjie Luo, Chang Liu, Di He, and Tie-Yan Liu. “Benchmarking graphormer on large-scale molecular modeling datasets”. arXiv preprint arXiv:2203.04810, 2022.

• Payal Bajaj, Chenyan Xiong, Guolin Ke, Xiaodong Liu, Di He, Saurabh Tiwary, Tie-Yan Liu, Paul Bennett, Xia Song, and Jianfeng Gao. “METRO: Efficient Denoising Pretraining of Large Scale Autoencoding Language Models with Model Generated Signals”. arXiv preprint arXiv:2204.06644, 2022.

• Yongchun Zhu, Fuzhen Zhuang, Jindong Wang, Guolin Ke, Jingwu Chen, Jiang Bian, Hui Xiong, Qing He. “Deep Subdomain Adaptation Network for Image Classification”. IEEE TNNLS 2021, 32(4), 1713–1722.

• Shuqi Lu, Di He, Chenyan Xiong, Guolin Ke, Waleed Malik, Zhicheng Dou, Paul Bennett, Tie-Yan Liu, and Arnold Overwijk. “Less Is More: Pretrain a Strong Siamese Encoder for Dense Text Retrieval Using a Weak Decoder”. EMNLP 2021.

• Shengjie Luo, Shanda Li, Tianle Cai, Di He, Dinglan Peng, Shuxin Zheng, Guolin Ke, Liwei Wang, and Tie-Yan Liu. “Stable, Fast and Accurate: Kernelized Attention with Relative Positional Encoding”. NeurIPS 2021.

• Chengxuan Ying, Mingqi Yang, Shuxin Zheng, Guolin Ke, Shengjie Luo, Tianle Cai, Chenglin Wu, Yuxin Wang, Yanming Shen, and Di He. “First Place Solution of KDD Cup 2021 & OGB Large-Scale Challenge Graph Prediction Track”. arXiv preprint arXiv:2106.08279, 2021.

• Chengxuan Ying, Tianle Cai, Shengjie Luo, Shuxin Zheng, Guolin Ke, Di He, Yanming Shen, and Tie-Yan Liu. “Do Transformers Really Perform Badly for Graph Representation?”. NeurIPS 2021.

• Dinglan Peng, Shuxin Zheng, Yatao Li, Guolin Ke, Di He, and Tie-Yan Liu. “How Could Neural Networks Understand Programs?”. ICML 2021.

• Chengxuan Ying, Guolin Ke, Di He, and Tie-Yan Liu. “LazyFormer: Self Attention with Lazy Update”. arXiv preprint arXiv:2102.12702, 2021.

• Guolin Ke, Di He, and Tie-Yan Liu. “Rethinking Positional Encoding in Language Pre-Training”. ICLR 2021.

• Qiyu Wu, Chen Xing, Yatao Li, Guolin Ke, Di He, and Tie-Yan Liu. “Taking Notes on the Fly Helps Language Pre-Training”. ICLR 2021.

• Mingqing Xiao, Shuxin Zheng, Chang Liu, Yaolong Wang, Di He, Guolin Ke, Jiang Bian, Zhouchen Lin, and Tie-Yan Liu. “Invertible Image Rescaling”. ECCV 2020.

• Zhenhui Xu, Guolin Ke, Jia Zhang, Jiang Bian, and Tie-Yan Liu. “Light Multi-Segment Activation for Model Compression”. AAAI 2020.

• Guolin Ke, Zhenhui Xu, Jia Zhang, Jiang Bian, and Tie-Yan Liu. “DeepGBM: A Deep Learning Framework Distilled by GBDT for Online Prediction Tasks”. KDD 2019.

• Guolin Ke, Qi Meng, Thomas Finley, Taifeng Wang, Wei Chen, Weidong Ma, Qiwei Ye, and Tie-Yan Liu. “LightGBM: A Highly Efficient Gradient Boosting Decision Tree”. NeurIPS 2017.

• Qi Meng, Guolin Ke, Taifeng Wang, Wei Chen, Qiwei Ye, Zhi-Ming Ma, and Tie-Yan Liu. “A Communication-Efficient Parallel Algorithm for Decision Tree”. NeurIPS 2016.

• Zhenhui Xu, Linyuan Gong, Guolin Ke, Di He, Shuxin Zheng, Liwei Wang, Jiang Bian, and Tie-Yan Liu. “MC-BERT: Efficient Language Pre-Training via a Meta Controller”. arXiv preprint arXiv:2006.05744, 2020.