PeopleofSuperintelligence

Researcher

Nan Duan

Nan Duan is an AI researcher who co-authored CodeBERT, a pretrained model for programming and natural languages, in 2020. He earned his B.S. in 2004 and his Ph.D. in 2011, both in computer science at Tianjin University. In 2021 he co-authored NÜWA, a Microsoft Research Asia paper, and in 2023 he co-authored Visual ChatGPT. In 2024 he co-authored Not All Tokens Are What You Need for Pretraining, a NeurIPS paper listed as a Best Paper Runner-up. As of October 2026, he is vice president at JD.COM and deputy director of Joy Future Academy, where he leads foundation model research, and he is a former Technical Fellow at StepFun.

Now at
JD.com
Works
332
Citations
16,456
h-index
57

Publication numbers from OpenAlex, as of 15 October 2026.

Timeline

  1. 2004–2011

    Computer science at Tianjin University

    Tianjin University confers his B.S. in 2004 and his Ph.D. in 2011, both in computer science.

  2. 2019

    A CCF-NLPCC honour

    In 2019, the CCF-NLPCC names him a Distinguished Young Scientist for his contributions to natural language processing.

  3. 2020

    The CodeBERT model

    He co-authors CodeBERT, a pretrained model for programming and natural languages, in 2020.

  4. November 2021

    The NÜWA visual model

    He co-authors NÜWA, a Microsoft Research Asia paper, in November 2021.

  5. 2022

    A CVPR best demo award

    His work receives the Best Demo Award at CVPR 2022.

  6. March 2023

    The Visual ChatGPT paper

    Visual ChatGPT, an arXiv paper posted in March 2023, lists him as a co-author.

  7. 2024

    A NeurIPS runner-up paper

    Not All Tokens Are What You Need for Pretraining, a 2024 NeurIPS paper listed as a Best Paper Runner-up, includes him as a co-author.

  8. As of October 2026

    Vice president at JD.COM

    He is vice president at JD.COM and deputy director of Joy Future Academy, where he leads foundation model research. He is a former Technical Fellow at StepFun.

Sources and credits

Photo: We have no published photo of Nan Duan yet. Send us a free-licensed photo.

  1. Towards AGI: Foundation Models for Language, Code, and Multimodality | Department of Data Science ds.cityu.edu.hk. Checked 15 October 2026.
  2. Nan Duan’s Homepage | May we journey far, yet return forever young. nanduan.github.io. Checked 15 October 2026.
  3. [2002.08155] CodeBERT: A Pre-Trained Model for Programming and Natural Languages arxiv.org. Checked 15 October 2026.
  4. [2111.12417] NÜWA: Visual Synthesis Pre-training for Neural visUal World creAtion arxiv.org. Checked 15 October 2026.
  5. NÜWA: Visual Synthesis Pre-training for Neural visUal World creAtion - Microsoft Research microsoft.com. Checked 15 October 2026.
  6. [2303.04671] Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models arxiv.org. Checked 15 October 2026.
  7. Not All Tokens Are What You Need for Pretraining proceedings.neurips.cc. Checked 15 October 2026.
  8. NeurIPS 2024 Awards neurips.cc. Checked 15 October 2026.