PeopleofSuperintelligence

Researcher

Xiaokang Chen

Also known as 陈小康

Xiaokang Chen is a DeepSeek AI researcher whose work links image understanding with image generation. He earned a computer science bachelor's degree and a PhD at Peking University, completed research internships at SenseTime, Microsoft Research Asia, Baidu Research and Shanghai Artificial Intelligence Laboratory, and joined DeepSeek in April 2024. He coauthored Janus and JanusFlow, DeepSeek-VL2 and DeepSeek-V3 in 2024, led the Janus series, and first-authored Janus-Pro in 2025. He received the WAIC Yunfan Award in July 2025 and, as of October 2026, is a technical staff member at DeepSeek who works on multimodal pre-training and post-training.

Now at
DeepSeek
Works
71
Citations
4,934
h-index
18
Badges

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 2015–2024

    Computer science at Peking University

    At Peking University, he earns a bachelor's degree in computer science in July 2019 and completes a PhD under Professor Gang Zeng in 2024.

  2. 2019–2023

    Research internships

    He works as a research intern at SenseTime Research from April 2019 to May 2020 under Kwan-Yee Lin and Wayne (Wenyan) Wu. At Microsoft Research Asia, he interns under Jingdong Wang from June 2020 to December 2021; at Baidu Research he works under Wang from December 2021 to December 2022, then joins Shanghai Artificial Intelligence Laboratory under Wenhai Wang and Jifeng Dai through November 2023.

  3. April 2024

    Joining DeepSeek

    In April 2024, he joins DeepSeek as an AGI researcher and works on multimodal pre-training and post-training for its large language models.

  4. December 2024

    DeepSeek-VL2 and DeepSeek-V3

    In December 2024, he coauthors DeepSeek-VL2, a mixture-of-experts vision-language model with dynamic tiling for high-resolution images. He also coauthors DeepSeek-V3, a mixture-of-experts language model with 671 billion total parameters and 37 billion activated for each token.

  5. January 2025

    First author of Janus-Pro

    In January 2025, he first-authors Janus-Pro, a larger Janus model with expanded training data and an optimized training strategy. He also serves as project lead and core contributor for the Janus series.

  6. 2025

    Janus and JanusFlow at CVPR

    In October 2024, he coauthors Janus, an autoregressive framework for multimodal understanding and generation. He coauthors JanusFlow in November 2024, which combines autoregressive language models with rectified flow; both papers appear in CVPR 2025. Janus separates visual encoders for understanding and generation.

  7. July 2025

    A WAIC Yunfan Award

    He receives the WAIC Yunfan Award in July 2025.

  8. As of October 2026

    DeepSeek multimodal research

    As of October 2026, he is a technical staff member at DeepSeek AI, driving multimodal pre-training and post-training for its large language models.

Career

  1. NowDeepSeekTechnical Staff Member
  2. 2022–2023Shanghai Artificial Intelligence LaboratoryResearch Intern
  3. 2021–2022BaiduResearch Intern
  4. 2020–2021Microsoft Research AsiaResearch Intern
  5. 2019–2020SenseTimeResearch Intern

Education

  1. 2024Peking UniversityPhD
  2. 2019Peking UniversityBachelor's degree

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Xiaokang Chen. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  2. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes Xiaokang Chen. We estimate that alumni of 17 major AI labs went on to found or lead 350 other organizations, including 226 they founded, and we trace 472 career moves directly between 23 labs since 2010.
  3. 16% of AI career moves in the 2020s went from a university to a companyIncludes Xiaokang Chen. We estimate that 16% of AI career moves in the 2020s went from a university to a company; 29% of moves starting at a university went to a company, and 76% ended the university role.
  4. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes DeepSeek. Our estimates of the organizations where 550 people worked directly before joining 9 frontier AI labs across 620 career transitions, comparing the main feeder organizations per lab.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

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

  1. Homepage - Welcome to Xiaokang Chen's Homepage charlescxk.github.io. Checked 21 October 2026.
  2. DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding arxiv.org. Checked 21 October 2026.
  3. DeepSeek-V3 Technical Report arxiv.org. Checked 21 October 2026.
  4. Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling arxiv.org. Checked 21 October 2026.
  5. Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation arxiv.org. Checked 21 October 2026.
  6. JanusFlow: Harmonizing Autoregression and Rectified Flow for Unified Multimodal Understanding and Generation arxiv.org. Checked 21 October 2026.
  7. CVPR 2025 Open Access Repository openaccess.thecvf.com. Checked 21 October 2026.
  8. CVPR 2025 Open Access Repository openaccess.thecvf.com. Checked 21 October 2026.