PeopleofSuperintelligence

Researcher

Rong Jin

Also known as 金榕

Rong Jin is a machine learning researcher whose work on visual search, forecasting and recommendation has moved from a university to Alibaba and Meta. A Carnegie Mellon computer science PhD (2003), he was a Michigan State University faculty member until he joined Alibaba in 2015. At Alibaba, he became a founding member of DAMO Academy and head of its Machine Intelligence Lab, coauthored the 2018 paper Visual Search at Alibaba on the Pailitao image-search product, and coauthored the FEDformer forecasting model in 2022. By 2023, he was affiliated with Meta, where he coauthored External Large Foundation Model, an ads recommendation framework, in 2025. He is at Meta, and his 2026 coauthored Mixture of Sequence framework targets long-sequence recommendation.

Now at
Meta
Badges

Timeline

  1. 2003

    A Carnegie Mellon doctorate

    He earns a PhD in computer science at Carnegie Mellon University in 2003, having studied at the Language Technologies Institute in the School of Computer Science.

  2. 2003–2015

    Faculty at Michigan State

    He is a faculty member in the Department of Computer Science and Engineering at Michigan State University until he joins Alibaba in 2015.

  3. 2015–2017

    Moving to Alibaba

    He joins Alibaba in 2015. In 2017, he is an invited speaker at IJCAI with the talk Deep Learning at Alibaba, which covers deep learning for recommender systems, transfer learning for multimedia data, DNN model compression and combinatorial optimization for a shipping packing problem.

  4. 2018

    Visual Search at Alibaba

    In 2018, he coauthors Visual Search at Alibaba at KDD, which describes the visual search algorithm and system behind Pailitao, the search-by-image product launched in 2014 that averages over 17 million daily active users in 2017.

  5. 2022

    The FEDformer forecasting model

    In 2022, he coauthors FEDformer with Tian Zhou, Ziqing Ma, Qingsong Wen, Xue Wang and Liang Sun, a Transformer that combines seasonal-trend decomposition with a frequency enhanced design for long-term series forecasting. On six benchmark datasets, it reduces prediction error by 14.8% for multivariate and 22.6% for univariate time series compared with earlier methods.

  6. May 2022

    Leaving DAMO Academy

    In May 2022, he is reported to leave Alibaba, where he is a vice president, associate dean of DAMO Academy, founding member of DAMO and head of its Machine Intelligence Lab. There he develops Pailitao, a visual search application that tens of millions of users use each day to search commodities with photos.

  7. By 2023

    Affiliated with Meta

    In November 2023, he coauthors Model-free Test Time Adaptation for Out-Of-Distribution Detection with YiFan Zhang, Xue Wang, Tian Zhou and others, with his affiliation given as Meta and the work done at Alibaba Group.

  8. As of October 2026

    Ads recommendation at Meta

    In 2025, he coauthors External Large Foundation Model with colleagues at AI at Meta, an ads recommendation framework in which a foundation-model-style teacher serves multiple smaller vertical models through external distillation. In March 2026, he coauthors Mixture of Sequence, a model-agnostic mixture-of-experts method for long-sequence recommendation, with his affiliation given as Meta.

Career

  1. NowMeta
  2. 2015–2022AlibabaVice President, Associate Dean, Founding Member
  3. 2003–2015Michigan State UniversityFaculty Member

Education

  1. 2003Carnegie Mellon UniversityPhD

From the biography.

  1. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes Rong Jin. 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. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Meta. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  3. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes Meta. 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.
  4. Of 943 people shaping AI who worked at Google, 446 work there nowIncludes Meta. Our estimates of where the 943 people shaping AI who worked at Google and Google DeepMind work now: 446 at Google and 452 across 278 other organizations.

Colleagues

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

Sources and credits

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

  1. Rong Jin's CMU homepage (After graduation version) cs.cmu.edu. Checked 22 October 2026.
  2. Weekly China AI News from May 16 to May 22 recodechinaai.substack.com. Checked 22 October 2026.
  3. Deep Learning at Alibaba | IJCAI ijcai.org. Checked 22 October 2026.
  4. [2102.04674] Visual Search at Alibaba arxiv.org. Checked 22 October 2026.
  5. Visual Search at Alibaba (ar5iv) ar5iv.labs.arxiv.org. Checked 22 October 2026.
  6. [2201.12740] FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecasting arxiv.org. Checked 22 October 2026.
  7. [2311.16420] Model-free Test Time Adaptation for Out-Of-Distribution Detection arxiv.org. Checked 22 October 2026.
  8. Model-free Test Time Adaptation for Out-Of-Distribution Detection (ar5iv) ar5iv.labs.arxiv.org. Checked 22 October 2026.
  9. External Large Foundation Model: How to Efficiently Serve Trillions of Parameters for Online Ads Recommendation arxiv.org. Checked 22 October 2026.
  10. Mixture of Sequence: Theme-Aware Mixture-of-Experts for Long-Sequence Recommendation arxiv.org. Checked 22 October 2026.