Feng Yu

I am Feng Yu, a PhD student in Computer Science at the University of Exeter, supervised by Prof. Jia Hu and Prof. Geyong Min. My current research focuses on LLM post-training, continual learning and adaptation, and efficient AI. I am particularly interested in how foundation models can be adapted efficiently while retaining capability under evolving tasks and distributions, and more recently in continual adaptation of agents under changing interfaces and workflows. My research is supported by China Scholarship Council and University of Exeter Scholarships.

Current interests

Post-training and efficient adaptation for large language models
Continual learning, retention, and robustness under evolving tasks and distributions
Evaluation and continual adaptation of agents under changing interfaces and workflows

Broader background

My broader background includes federated learning, continual learning, efficient AI, privacy-aware learning systems, and earlier work on deep reinforcement learning, blockchain-enabled learning systems, and engineering-oriented automation.

I am open to discussing research ideas and academic collaborations. Feel free to reach out by email.

News

May 18, 2025	One research paper FedTaLoRA has been released!
Nov 20, 2024	The first comprehensive survey FCL for Edge-AI has been released!
Sep 1, 2023	Started my PhD at Exeter!

Selected publications

2025

Blockwise Hadamard high-Rank Adaptation for Parameter-Efficient LLM Fine-Tuning

Feng Yu , Jia Hu, and Geyong Min

arXiv preprint arXiv:2509.21637 2025

Bib HTML
@article{yu2025blockwise, title = {Blockwise Hadamard high-Rank Adaptation for Parameter-Efficient LLM Fine-Tuning}, author = {Yu, Feng and Hu, Jia and Min, Geyong}, journal = {arXiv preprint arXiv:2509.21637}, year = {2025}, bibtex_show = {true}, selected = {true}, abbr = {arXiv}, html = {https://arxiv.org/abs/2509.21637} }
Efficient Federated Class-Incremental Learning of Pre-Trained Models via Task-agnostic Low-rank Residual Adaptation

Feng Yu , Jia Hu, and Geyong Min

arXiv preprint arXiv:2505.12318 2025

Bib HTML
@article{yu2025efficient, title = {Efficient Federated Class-Incremental Learning of Pre-Trained Models via Task-agnostic Low-rank Residual Adaptation}, author = {Yu, Feng and Hu, Jia and Min, Geyong}, journal = {arXiv preprint arXiv:2505.12318}, year = {2025}, bibtex_show = {true}, selected = {true}, abbr = {arXiv}, html = {https://arxiv.org/abs/2505.12318} }

2024

Federated Continual Learning for Edge-AI: A Comprehensive Survey

Zi Wang, Fei Wu, Feng Yu , Yurui Zhou, Jia Hu, and Geyong Min

arXiv preprint arXiv:2411.13740 2024

Bib HTML
@article{zi2024federated, title = {Federated Continual Learning for Edge-AI: A Comprehensive Survey}, author = {Wang, Zi and Wu, Fei and Yu, Feng and Zhou, Yurui and Hu, Jia and Min, Geyong}, journal = {arXiv preprint arXiv:2411.13740}, year = {2024}, bibtex_show = {true}, selected = {true}, abbr = {arXiv}, html = {https://arxiv.org/abs/2411.13740} }