Academic profile
Gaozhuo Liu
Undergraduate researcher working on federated learning, trustworthy LLMs and secure agent systems.
About
I am an undergraduate student in Telecommunication Engineering at Beijing University of Posts and Telecommunications and a researcher with the Intelligent and Distributed Computing Lab at Huazhong University of Science and Technology.
My work focuses on distributed, adaptive and trustworthy learning systems, with particular interest in federated continual learning, safety-aware LLM adaptation and secure multimodal agent collaboration.
Education
Beijing University of Posts and Telecommunications
B.Sc. (Engineering), Telecommunication Engineering
Average mark 87.93 / 100 · GPA 3.59 / 4.0 · Top 20%
Research
Intelligent and Distributed Computing Lab, HUST
Wuhan, China · Working closely with Yichen Li
Research on heterogeneous federated continual learning, lightweight incremental learning, personalized federated LLM safety, poison-resilient agent skill evolution and multimodal latent-channel security.
Published work
Lightweight Federated Incremental Learning via Decoupled Replay
Proceedings of the International Conference on Machine Learning (ICML), 2026.
Only formally published work is shown publicly. Other manuscripts will be added after final decisions.
Interests
Skills
Programming: Python, C/C++, MATLAB, Bash, LaTeX
ML systems: PyTorch, Transformers, LoRA/PEFT, distributed training, Docker, CUDA
Languages: Chinese (native), English (CEFR B2)