Research Profile
PhD Candidate · University of Glasgow
Hover or focus the photo to expand it.
Federated Learning · Edge Systems · LLM Fine-tuning
Ke Xiao
PhD Candidate at the University of Glasgow. I design resilient distributed AI systems for federated and edge environments.
Latest News
Apr 30, 2026
Publication
Paper accepted to ICML 2026!
FedHera: Towards Drift-Resilient Federated Fine-tuning with Heterogeneous Resources.
View TimelineFeatured Research
Latest Selected Paper
FedHera: Towards Drift-Resilient Federated Fine-tuning with Heterogeneous Resources
Abstract
FedHera is a resource-decoupled federated fine-tuning framework for large language models under heterogeneous client bandwidth and compute constraints.
Research Archive
Selected Publications
A curated view of journal and conference work.
2025
Selected
FLgym: Towards Robust and Byzantine-Resilient Federated Learning
IEEE Transactions on Information Forensics and Security (T-IFS)
2025
Selected
Resilient Inference for Personalized Federated Learning in Edge Environments
45th IEEE International Conference on Distributed Computing Systems (ICDCS 2025)
2024
Selected
Making Serverless Not So Cold in Edge Clouds: A Cost-Effective Online Approach
IEEE Transactions on Mobile Computing (TMC)
Timeline
News
Recent updates & milestones
Paper accepted to ICML 2026!
FedHera: Towards Drift-Resilient Federated Fine-tuning with Heterogeneous Resources.
Paper accepted to IEEE T-IFS!
FLgym: Towards Robust and Byzantine-Resilient Federated Learning.
Best Runner-Up Student Paper Award at IEEE ICDCS 2025!
Received for research on Byzantine-resilient federated learning.
Paper accepted to IEEE ICDCS 2025!
Unified Parallel Semantic Log Parsing based on Causal Graph Construction for Attack Attribution.
Profile
About & Education
I am a PhD candidate in the KDES Group at the University of Glasgow, working on resilient distributed AI systems.
My research sits at the intersection of federated learning, edge computing, and large language model fine-tuning. I focus on making learning systems more adaptive, reliable, and secure under resource and threat heterogeneity.
Current Focus
- Byzantine-resilient federated learning
- Heterogeneous fine-tuning for large language models
- Robust inference and scheduling in edge environments
Research Interests
Federated Learning
Edge Computing
LLM Fine-tuning
Distributed AI
System Security
Academic Service
Reviewer for journals including IEEE TMC, Computer Communications, and Computer Networks.
Education
2024 - Present
PhD Candidate
University of Glasgow
2021 - 2024
Master of Computer Technology
Beijing Institute of Technology
2017 - 2021
Bachelor of Software Engineering
Dalian University of Technology