Yang Fan（范阳）

Personal Information

Personal Photo

Senior Algorithm Engineer DAMO Academy, Alibaba Group

Building 2, No. 5 DanLing Street, Haidian District, Beijing, China.

Bio

Yang Fan is a Senior Algorithm Engineer of DAMO Acadamy, Alibaba Group. He has a wide range of research interests in artificial intelligence, including large language models, deep learning, natural language processing, learning to teach, AI for medicine, etc.
You can refer to the personal page: https://fyabc.github.io/.

Education

Sep. 2017 - Jun. 2022: School of Computer Science and Technology, University of Science and Technology of China, Doctor of Philosophy (Ph.D.)
Sep. 2013 - Jun. 2017: School of Computer Science and Technology, University of Science and Technology of China, Bachelor of Science (B.S.)

Experiences

Jul. 2022 - Now: Senior Algorithm Engineer, DAMO Academy, Alibaba Group.
Jul. 2018 - Jun. 2022: Research Intern, Machine Learning Group, Microsoft Research Asia, Mentor: Prof. Tao Qin
Jul. 2016 - Jun. 2017: Research Intern, Machine Learning Group, Microsoft Research Asia, Mentor: Prof. Tao Qin

Research Interests

Large Language Models (LLMs): Training, fine-tuning and data engineering of large language models.
Learning to Teach: A general meta-learning framework to automatically guide the training of AI tasks with a teacher model.
Neural Architecture Search: Automatically discover better architectures for AI tasks, especially for natural language processing tasks.
Neural Machine Translation: Train a deep neural network to translate from on language to another. Focusing more on real-world tasks.
AI for Medicine: Use AI and deep learning technologies to assist in drug discovery and disease treatment.

Honors & Awards

Jun. 2021, 6th place in the PCQM4M-LSC dataset prediction task
In the first OGB Large-Scale Challenge (OGB-LSC)
[Leaderboard]
Aug. 2019, championship on 8 translation tasks
In ACL 2019 fourth conference on machine translation (WMT19)
[PDF] [News] [Leaderboard]
2020, Huawei Scholarship
2017, Bao Gang Education Scholarship

2015, National Scholarship for Encouragement
2014, Individual Scholarship

Publications

Learning to Reweight with Deep Interactions [PDF]
Yang Fan, Yingce Xia, Lijun Wu, Shufang Xie, Weiqing Liu, Jiang Bian, Tao Qin, Xiang-Yang Li
The 35th AAAI Conference on Artificial Intelligence (AAAI-2021)
Searching Better Architectures for Neural Machine Translation [PDF]
Yang Fan, Fei Tian, Yingce Xia, Tao Qin, Xiang-Yang Li, Tie-Yan Liu
IEEE Transactions on Audio, Speech and Language Processing (TASLP), 18 May 2020, Volume 28, Pages 1574-1585
Learning to Teach [PDF]
Yang Fan, Fei Tian, Tao Qin, Xiang-Yang Li, Tie-Yan Liu
The Sixth International Conference on Learning Representations (ICLR-2018)
Multi-branch Attentive Transformer [PDF] [link]
Yang Fan, Shufang Xie, Yingce Xia, Lijun Wu, Tao Qin, Xiang-Yang Li, Tie-Yan Liu
arXiv, 2020.
Back Translation for Molecule Generation [PDF] [link]
Yang Fan, Yingce Xia, Jinhua Zhu, Lijun Wu, Shufang Xie, Tao Qin
Bioinformatics, 1 March 2022, Volume 38, Issue 5, Pages 1244-1251
Microsoft Research Asia’s Systems for WMT19 [PDF]
Yingce Xia, Xu Tan, Fei Tian, Fei Gao, Weicong Chen, Yang Fan, Linyuan Gong, Yichong Leng, Renqian Luo, Yiren Wang, Lijun Wu, Jinhua Zhu, Tao Qin, and Tie-Yan Liu
The Fourth Conference on Machine Translation (WMT-2019)
Learning to Teach with Dynamic Loss Functions [PDF]
Lijun Wu, Fei Tian, Yingce Xia, Yang Fan, Tao Qin, Jianhuang Lai, Tie-Yan Liu
Proceedings of the Thirty-second International Conference on Neural Information Processing Systems (NIPS-2018)
Sequence Generation with Mixed Representations [PDF]
Lijun Wu, Shufang Xie, Yingce Xia, Yang Fan, Tao Qin, Jianhuang Lai, Tie-Yan Liu
The Thirty-seventh International Conference on Machine Learning (ICML-2020)
End-to-end Entity-Aware Neural Machine Translation [PDF] [link]
Shufang Xie, Yingce Xia, Lijun Wu, Yiqing Huang, Yang Fan, Tao Qin
Machine Learning 2022 (ML-2022)
Learning to Teach for Function Space
Yang Fan, Fei Tian, Yingce Xia, Lijun Wu, Tao Qin, Xiang-Yang Li

Programming Skills

Programming Languages: Master user of Python, frequent user of C++ and LaTeX
Deep Learning Tools: Master user of PyTorch, frequent user of Tensorflow