Zihuan Qiu (邱子欢)

I am currently a third-year Ph.D. student at the School of Information and Communication Engineering, University of Electronic Science and Technology of China (UESTC), supervised by Prof. Fanman Meng. I received my M.S. from UESTC in 2023.

My research centers on continual learning, broadly understood as enabling models to continuously learn, adapt, and accumulate knowledge over time. Around this goal, (1) continual learning for neural networks, where I focus on reducing catastrophic forgetting and balancing the acquisition of new knowledge with the preservation of old knowledge; (2) model parameter merging, which studies how to efficiently combine multiple task-specific models into a unified model so that knowledge learned from different tasks can be reused without costly retraining; and (3) agent self-evolution, where I explore how LLM and VLM agents can develop reusable skills, maintain long-term memory, and adapt to open-ended environments through tool use and recursive learning.

Email: zihuanqiu AT qq DOT com

Google Scholar GitHub Email