↑

Chunlin Tian

Ph.D. Student · University of Macau
Visiting Ph.D. Student · University of Cambridge
Email: yc27402@um.edu.mo / ct720@cam.ac.uk
🚀 Open to: Faculty · Postdoc · Research Scientist
Chunlin Tian

Short Bio

👋 I am a final-year Ph.D. candidate in Computer Science at the State Key Laboratory of Internet of Things for Smart City (SKL-IOTSC), University of Macau, advised by Prof. Li Li and Prof. Chengzhong Xu (IEEE Fellow). I am currently a Visiting Researcher at the Machine Learning Systems Group, University of Cambridge, working with Prof. Nicholas D. Lane. I also collaborate with Dr. Zhan Shi and Prof. Zhijiang Guo. My research focuses on Efficient Machine Learning System and Edge AI, I specialize in software–hardware co-design, spanning from algorithms to systems, to address the stringent SWaP (Space, Weight, and Power) constraints in deploying foundation models on edge devices. I have published 20+ papers in top-tier journals and conferences, including MICRO, USENIX ATC, TPDS, NeurIPS, and ICML, with several selected for Oral presentations.

Education

Selected Publications

For a complete list of my publications, please refer to my Google Scholar profile.

ATC 2025
CLONE: Customizing LLMs for Efficient Latency-Aware Inference at the Edge.
Chunlin Tian, Xinpeng Qin, Kahou Tam, Li Li, Zijian Wang, Yuanzhe Zhao, Minglei Zhang and Chengzhong Xu
ATC 2025. [AR: 100/634 = 15.8%]
“When one size doesn’t fit all, CLONE it to beat the edge latency!”
NeurIPS 2024
HydraLoRA: An Asymmetric LoRA Architecture for Efficient Fine-Tuning. Oral
Chunlin Tian, Zhan Shi, Zhijiang Guo, Li Li and Chengzhong Xu
NeurIPS 2024. [AR: 61/15671 = Top 0.4%]
“Unleash the asymmetric heads of Hydra to tame the beast of fine-tuning!”
ICML 2024
Ranking-based Client Imitation Selection for Efficient Federated Learning.
Chunlin Tian, Zhan Shi, Xinpeng Qin, Li Li and Chengzhong Xu
ICML 2024. [AR: 2609/9653 = 27.0%]
“Don’t play dice with clients; Rank and Imitate for the win!”
TPDS 2024
Breaking the Memory Wall for Heterogeneous Federated Learning via Model Splitting.
Chunlin Tian, Li Li, Kahou Tam, Yebo Wu, Chengzhong Xu
IEEE TPDS 2024.
“Smash the Memory Wall by splitting the model, not the device!”
TPDS 2026 Floe
Floe: Federated Specialization for Real-Time LLM-SLM Inference.
Chunlin Tian, Kahou Tam, Yebo Wu, Shuaihang Zhong, Li Li, Nicholas D. Lane, Chengzhong Xu
TPDS 2026.
“Specialize federated LLM-SLM pipelines to keep real-time edge inference afloat!”
IJCAI 2024
FedGCS: A Generative Framework for Efficient Client Selection in Federated Learning via Gradient-based Optimization.
Zhiyuan Ning*, Chunlin Tian*, Meng Xiao, Wei Fan, Pengyang Wang, Li Li, Pengfei Wang and Yuanchun Zhou
IJCAI 2024. [AR: 791/5651 = 14.0%]
“Why search for the perfect client when you can generate one via gradients!”
TMLR 2026
A Survey on Federated Fine-Tuning of Large Language Models.
Yebo Wu, Chunlin Tian, Jingguang Li, He Sun, KaHou Tam, Zhanting Zhou, Haicheng Liao, Jing Xiong, Zhijiang Guo, Li Li, Chengzhong Xu
TMLR 2026.
“A practical map for navigating federated fine-tuning in the LLM era!”
ICLR 2023 Tiny
Learn to Select: Efficient Cross-device Federated Learning via Reinforcement Learning.
Chunlin Tian, Zhan Shi and Li Li
ICLR 2023, Tiny.
“Don’t just train every client, learn which ones are worth the round!”
SenSys 2024
FedHybrid: Breaking the Memory Wall of Federated Learning via Hybrid Tensor Management.
Kahou Tam, Chunlin Tian, Li Li, Haikai Zhao, Chengzhong Xu
SenSys 2024. [AR: 58/313 = 18.5%]
“When memory gets tight, go Hybrid to keep the training flowing!”
MICRO 2022
HARMONY: Heterogeneity-Aware Hierarchical Management for Federated Learning System.
Chunlin Tian, Li Li, Zhan Shi, Jun Wang and Chengzhong Xu
MICRO 2022. [AR: 83/369 = 22.5%]
“Bring order to heterogeneous chaos with hierarchical HARMONY!”

Professional Services

Mentored Students

✨If you are passionate about research and innovation, I warmly invite you to join us on this exciting academic journey!

Awards

Teaching Assistants

Visitor map