Full Publications

indicates equal contribution, indicates members of the research group I lead.
[IEEE TPDS'25] "Co-designing Transformer Architectures for Distributed Inference with Low Communication".IEEE Transactions on Parallel and Distributed Systems (TPDS), 2025. CCF-A
[INFOCOM'25] "Jupiter: Fast and Resource-Efficient Collaborative Inference of Generative LLMs on Edge Devices".IEEE International Conference on Computer Communications (INFOCOM), 2025. CCF-A
[IEEE MSN'24] "MIX3D: A Mixed Representation for Communication-Efficient Distributed 3DGS Training".The 20th IEEE International Conference on Mobility, Sensing and Networking (IEEE MSN), 2024. CCF-C
[IFIP NPC'24] "MEGA: Mesh-Aligned 3DGS Towards Geometry-Preserving Online Reconstruction".IFIP International Conference on Network and Parallel Computing (IFIP NPC), 2024. CCF-C
🏵️ Best Student Paper Award
[ArXiv'24] "Edge Graph Intelligence: Reciprocally Empowering Edge Networks with Graph Intelligence".Under review
[ICPP'24] "Pluto and Charon: A Time and Memory Efficient Collaborative Edge AI Framework for Personal LLMs Fine-Tuning".International Conference on Parallel Processing (ICPP), 2024. CCF-B
[IEEE WCM'24] "Implementation of Big AI Models for Wireless Networks with Collaborative Edge Computing".IEEE Wireless Communications (IEEE WCM), 2024. 中科院一区(Top) IF=12.9
[INFOCOM'24] "Galaxy: A Resource-Efficient Collaborative Edge AI System for In-situ Transformer Inference".IEEE International Conference on Computer Communications (INFOCOM), 2024. CCF-A
🏵️ IEEE ComSoc Student Travel Grant Award
[MobiCom'24] "Asteroid: Resource-Efficient Hybrid Pipeline Parallelism for Collaborative DNN Training on Heterogeneous Edge Devices".Annual International Conference On Mobile Computing And Networking (MobiCom), 2024. CCF-A
🏵️ Top #1 Conference in Mobile Computing and Computer Networks
[DATE'24] "Communication-Efficient Model Parallelism for Distributed In-situ Transformer Inference".Design, Automation and Test in Europe Conference (DATE), 2024. CCF-B
[ICPP'22] "Eco-FL: Adaptive Federated Learning with Efficient Edge Collaborative Pipeline Training".International Conference on Parallel Processing (ICPP), 2022. CCF-B