PeopleofSuperintelligence

Researcher

Yifan Bai

AI researcher Yifan Bai works across autoregressive visual tracking, autonomous driving and agentic language models. They studied computer science at Xidian University and software engineering at Xi’an Jiaotong University, then coauthored ARTrack, ARTrackV2, ARPTrack and FutureSightDrive. At Megvii and Alibaba DAMO Academy, Bai worked on autonomous driving and embodied intelligence before coauthoring the Kimi K2, Kimi K2.5 and Kimi K3 technical reports. At Alibaba DAMO Academy, they work as a researcher, a role they have held since June 2025.

Now at
Alibaba
Badges

Timeline

  1. 2018–2025

    Bachelor’s and master’s studies

    At Xidian University, Bai studies computer science from 2018 to 2022 and software engineering at Xi’an Jiaotong University from 2022 to 2025, with Xing Wei as adviser.

  2. 2023–2024

    ARTrack and ARTrackV2

    In 2023, the ARTrack paper casts object tracking as a coordinate-sequence task, and Bai is a coauthor. In 2024, ARTrackV2 models an object’s trajectory and appearance in one autoregressive framework.

  3. December 2023–June 2025

    Megvii and Alibaba DAMO internships

    At Megvii Technology, Bai interns in autonomous driving from December 2023 to May 2024. At Alibaba DAMO Academy’s Visual Technology Center, they work on embodied intelligence from May 2024 to June 2025.

  4. 2025

    ARPTrack adds pretraining

    At CVPR 2025, Bai coauthors ARPTrack with Shiyi Liang, Yihong Gong and Xing Wei. The system adds unsupervised autoregressive video pretraining to ARTrackV2 to model changes in object appearance and motion.

  5. 2025

    Visual reasoning for autonomous driving

    The 2025 FutureSightDrive paper lists Bai among its authors and describes a visual spatio-temporal chain-of-thought system for autonomous-driving VLA models.

  6. July 2025

    Kimi K2 technical report

    The July 2025 Kimi K2 technical report lists Bai as an author and describes a 1-trillion-parameter mixture-of-experts model with 32 billion active parameters.

  7. February–July 2026

    Kimi K2.5 and K3

    Kimi K2.5, released in February 2026, is an open-source multimodal agent model trained jointly on text and vision, and Bai is a coauthor. Kimi K3, published in July 2026, is a native-vision model with a one-million-token context window, with Bai among its coauthors.

  8. Since June 2025

    Researcher at Alibaba DAMO Academy

    Alibaba DAMO Academy lists Bai as a researcher starting in June 2025. Their work includes embodied intelligence, autonomous driving, multimodal models and visual generation.

Career

  1. Since 2024AlibabaIntern, Researcher
  2. 2023–2024MegviiIntern

Education

  1. 2025Xi'an Jiaotong UniversityMaster's degree
  2. 2022Xidian UniversityBachelor's degree

From the biography.

  1. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes Alibaba. 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.
  2. Universities (31%) and startups and other companies (25%) employ the largest shares of 2020–2026 entrantsIncludes Alibaba. Among 819 people whose first recorded role started in 2020–2026 and who have a known current organization, 60% work in commercial sectors (25% at startups and other companies, 23% at frontier labs and 13% at big tech, with 9 in two), and 31% work at universities, compared with 66% for earlier entrants.
  3. The top 10 organizations employed 22% of people shaping AI in 2025: 51% of company researchers and 21% of university researchersIncludes Alibaba. We estimate the share of people shaping AI who worked at the top 10 organizations each year from 1955 to 2025, comparing concentration across companies, universities, and institutes and labs.
  4. 16% of AI career moves in the 2020s went from a university to a companyIncludes Alibaba. 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.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

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

  1. AlexDotHam (Yifan Bai) · GitHub github.com. Checked 21 October 2026.
  2. Yifan Bai | alphaXiv alphaxiv.org. Checked 21 October 2026.
  3. Autoregressive Visual Tracking openaccess.thecvf.com. Checked 21 October 2026.
  4. ARTrackV2: Prompting Autoregressive Tracker Where to Look and How to Describe artrackv2.github.io. Checked 21 October 2026.
  5. Autoregressive Sequential Pretraining for Visual Tracking openaccess.thecvf.com. Checked 21 October 2026.
  6. FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Driving miv-xjtu.github.io. Checked 21 October 2026.
  7. Kimi K2: Open Agentic Intelligence arxiv.org. Checked 21 October 2026.
  8. README.md · MoonshotAI/Kimi-K2 · GitHub github.com. Checked 21 October 2026.
  9. Kimi K2.5: Visual Agentic Intelligence arxiv.org. Checked 21 October 2026.
  10. Kimi K3: Open Frontier Intelligence arxiv.org. Checked 21 October 2026.