Qitao Tan

Ph.D. Student at the University of Georgia

1.png

Hi! I am Qitao Tan. I am currently a second year Ph.D. student in School of Computing at the University of Georgia, advised by Dr.Geng Yuan. I received my B.E. degree in Information Science from Central China Normal University, China, in 2024.

My research focuses on efficient AI, with an emphasis on the training, deployment, and acceleration of large-scale deep neural networks (e.g., LLMs, multi-modal models, CNNs), and hardware-software co-design for edge intelligence. My research targets diverse platforms, ranging from powerful cloud clusters to resource-constrained edge devices.

News

09/2025 One paper is accepted by Neurips 2025.
08/2025 One paper is selected as one of the ICCAD 2025 Best Paper Nominees !
06/2025 One paper is accepted by ICCAD 2025.
04/2025 One collaboration paper is accepted by GLSVLSI 2025.
01/2025 One paper is accepted by ICLR 2025.
09/2024 One collaboration paper is accepted by ASP-DAC 2025.
08/2024 Join the University of Georgia, School of computing as a PhD student !

Selected publications

  1. Neurips
    Harmony in divergence: Towards fast, accurate, and memory-efficient zeroth-order llm fine-tuning
    Qitao Tan, Jun Liu, Zheng Zhan, Caiwei Ding, Yanzhi Wang, Xiaolong Ma, Jaewoo Lee, Jin Lu, and Geng Yuan
    In Advances in Neural Information Processing Systems, 2025
    HTML
  2. ICCAD
    Perturbation-efficient zeroth-order optimization for hardware-friendly on-device training
    Qitao Tan, Sung-En Chang, Rui Xia, Huidong Ji, Chence Yang, Ci Zhang, Jun Liu, Zheng Zhan, Zhenman Fang, Zhuo Zou, Yanzhi Wang, Jin Lu, and Geng Yuan
    In International Conference On Computer Aided Design (Best Paper Nomination), 2025
    HTML
  3. ICLR
    Mutual Effort for Efficiency: A Similarity-based Token Pruning for Vision Transformers in Self-Supervised Learning
    Sheng Li*, Qitao Tan*, Yue Dai, Zhenglun Kong, Tianyu Wang, Jun Liu, Ao Li, Ninghao Liu, Yufei Ding, Xulong Tang, and Geng Yuan
    In The Thirteenth International Conference on Learning Representations, 2025
    HTML