Chenhui Zhang

Research Engineer, Google DeepMind

prof_pic.jpg

I am a Research Engineer at Google DeepMind, working on AI for Science.

I graduated from the MIT Institute for Data, Systems, and Society (IDSS), where I was fortunate to be advised by Prof. Sherrie Wang. I was part of the Laboratory for Information and Decision Systems and the Earth Intelligence Lab, where we try to democratize the access to geospatial machine learning and better understand our planet.

Prior to MIT, I obtained my B.S. degree in Computer Science at the University of Illinois Urbana-Champaign, where I worked on various research problems in remote sensing, sustainable agriculture, and trustworthy machine learning.

My long-term research agenda aims to build accurate, efficient, and trustworthy machine learning systems for climate and geospatial science with a low barrier of entry. I hope the research products we build can address the pressing issues in sustainable agriculture, land management, and agroecosystem modeling, thereby contributing to our collective efforts to mitigate and adapt to climate change.

selected publications

  1. alphaearth.png
    AlphaEarth Foundations: An Embedding Field Model for Accurate and Efficient Global Mapping from Sparse Label Data
    Christopher F Brown, Michal R Kazmierski, Valerie J Pasquarella, William J Rucklidge, Masha Samsikova, Chenhui Zhang , and 13 more authors
    arXiv Preprint, 2025
  2. vleo-teaser.png
    Good at Captioning, Bad at Counting: Benchmarking GPT-4V on Earth Observation Data
    Chenhui Zhang, and Sherrie Wang
    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2024
  3. decodingtrust.png
    DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models
    Boxin Wang*Weixin Chen*, Hengzhi Pei*, Chulin Xie*Mintong Kang*Chenhui Zhang* , and 13 more authors
    Thirty-seventh Conference on Neural Information Processing Systems Datasets and Benchmarks Track, Outstanding Paper Award, 2023
    * indicates equal contribution
  4. cross-scale-tillage.jpg
    Cross-scale Sensing of Field-Level Crop Residue Cover: Integrating Field Photos, Airborne Hyperspectral Imaging, and Satellite Data
    Sheng WangKaiyu GuanChenhui Zhang, Qu Zhou, Sibo Wang, Xiaocui Wu , and 9 more authors
    Remote Sensing of Environment, 2023