Ci Zhang

I am Ci Zhang, a second-year Ph.D. student in Computer Science at the University of Georgia (UGA), advised by Geng Yuan. Before that, I got my M.S. degree at the Ohio State University, supervised by Rongjun Qin. My research interests include MLLM/LLM and generative models. Currently, I am focusing on diffusion models and continual learning on MLLM/LLM. Feel free to contact me if my work aligns with your team’s goals!

Research Interests

My research focuses on:

  • Continual learning for MLLM/LLM
  • Diffusion models and their applications
  • Efficient VLA

More broadly, I am interested in efficient and robust learning algorithms that allow modern generative models to adapt to new data and tasks over time without catastrophic forgetting.

News

02/2026    One paper is accepted to CVPR 2026.

06/2025    One paper is accepted to ICCAD 2025 (Best Paper Nominees).

06/2025    One paper is accepted to GLSVLSI 2025 (Oral).

03/2025    One paper is accepted to IJCAI 2025.

Selected Publication

CVPR'26 CVPR 2026 paper figure
Roots Beneath the Cut: Uncovering the Risk of Concept Revival in Pruning-Based Unlearning for Diffusion Models
Ci Zhang, Zhaojun Ding, Chence Yang, Jun Liu, Xiaoming Zhai, Shaoyi Huang, Beiwen Li, Xiaolong Ma, Jin Lu, Geng Yuan
The IEEE/CVF Conference on Computer Vision and Pattern Recognition 2026, (CVPR 2026)
GLSVLSI'25 GLSVLSI 2025 paper figure
Towards Memory-Efficient and Sustainable Machine Unlearning on Edge using Zeroth-Order Optimizer
Ci Zhang, Chence Yang, Qitao Tan, Jun Liu, Ao Li, Yanzhi Wang, Jin Lu, Jinhui Wang, Geng Yuan
Great Lakes Symposium on VLSI 2025, (GLSVLSI 2025) (Oral)
Under Review Physiological time signals imputation paper figure
Advancing Missing Data Reconstruction and Imputation for Physiological Time Signals via Mixture of Receptive Fields and Multi-Denoising-Factor Fusion
Ci Zhang, Huayu Li, Changdi Yang, Yanzhi Wang, Xiaolong Ma, Jin Lu, Ao Li, Geng Yuan
Under Review