Xinzi Cao

prof_pic.jpg

Xinzi Cao is an Research Assistant at the Institute of Embodied Intelligence, Pengcheng Laboratory. She received his Ph.D. from Sun Yat-sen University and Pengcheng Laboratory, under the supervision of Prof. Yutong Lu. His research focuses on open-world intelligence, aiming to develop AI systems that can perceive, learn, and adapt to novel and evolving environments. His previous work spans generalized category discovery, continual learning, and multimodal learning, enabling models to discover new concepts and acquire knowledge beyond predefined categories. She also explores large language models for AI systems, including AscendKernelGen, which leverages LLMs to automate NPU kernel generation. His current research focuses on embodied intelligence and multimodal learning for interaction with the physical world.

news

Apr 14, 2026 One papers is accepted by ACL Findings 🎉.
Mar 18, 2026 Two papers are accepted by ICME 2026 😀.
Jan 12, 2026 One Technical Report on NPU Kernel generation is released in arxiv.
Dec 04, 2025 One paper on prototypes debiasing is accepted by IEEE TCSVT 🎉🎉.
Jun 26, 2025 One paper on Prompt Tunning is accepted by the journal ICCV 2025 🎉🎉.

selected publications

  1. ascendKernelGen.png
    AscendKernelGen: A Systematic Study of LLM-Based Kernel Generation for Neural Processing Units
    Xinzi Cao, Jianyang Zhai, Pengfei Li, and 17 more authors
    2026
  2. debiasgcd.png
    Margin-Aware Prototype Debiasing for Generalized Category Discovery
    Xinzi Cao, Feidiao Yang, Xiawu Zheng, and 3 more authors
    IEEE Transactions on Circuits and Systems for Video Technology, 2025
  3. allgcd.png
    AllGCD: Leveraging All Unlabeled Data for Generalized Category Discovery
    Xinzi Cao, Ke Chen, Feidiao Yang, and 3 more authors
    In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Oct 2025
  4. STP.png
    Efficient Event Camera Data Pretraining with Adaptive Prompt Fusion
    Quanmin Liang, Qiang Li, Shuai Liu, and 6 more authors
    In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Oct 2025
  5. legogcd.png
    Solving the Catastrophic Forgetting Problem in Generalized Category Discovery
    Xinzi Cao, Xiawu Zheng, Guanhong Wang, and 5 more authors
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
  6. locloc.png
    LocLoc: Low-level Cues and Local-area Guides for Weakly Supervised Object Localization
    Xinzi Cao, Xiawu Zheng, Yunhang Shen, and 4 more authors
    In Proceedings of the 31st ACM International Conference on Multimedia (ACMMM), 2023