Jun Xia is currently at the Hong Kong University of Science and Technology and the ACCESS Chip Center as a Postdoctoral Research Fellow, advised by Prof. Yuan Xie. He has been a Postdoctoral Research Associate at the University of Notre Dame since 2023 (advised by Prof. Yiyu Shi). He obtained his Ph.D. degree from East China Normal University in 2023, advised by Prof. Mingsong Chen. He obtained his M.S. degree and B.S. degree from Jiangnan University (advised by Prof. Zhilei Chai and Prof. Wei Yan from Peking University) and Hainan University in 2019 and 2016, respectively.
His research interests lie in the intersection of Computer Architecture, Embedded Systems, and AI, with a current focus on Cloud-Edge Computing Architecture, Embodied AI Systems and On-Device Medical AI & Edge Healthcare.
๐ Selected Publications
๐ฅ News
- 2026.10 ๐๐ A paper was accepted by TMI.
- 2026.10 ๐๐ A paper was accepted by BIBM.
- 2026.09 ๐๐ Elected as an executive committee member of CCF Embedded Systems Committee (CCF TCEBS).
- 2026.09 ๐๐ Invited as a session chair at ICCD 2026.
- 2026.07 ๐๐ A paper was accepted by RTSS 2026.
- 2026.05 ๐๐ A paper was accepted by ICML 2026 (Congratulations to Yuejun).
- 2026.02 ๐๐ A paper was accepted by DAC 2026.
- 2026.02 ๐๐ Two papers were accepted by CVPR 2026 (Congratulations to Boyu and Zixuan).
๐ Selected Publications
RTSS 2026
CCF-A
Acceptance Ratio: 13%
Embedded System
ICML 2026
CCF-A
Acceptance Ratio: 25%
Machine Learning
DAC 2026
CCF-A
Acceptance Ratio: 25%
Embedded System
We introduce RTFL, a novel energy-aware Real-Time Federated Learning framework based on Multi-Agent Reinforcement Learning (MARL), aiming to enhance the knowledge sharing across AIoT devices within a specified training time constraint. Specifically, RTFL employs an Adaptive Quantization-based Multi-Agent Scheduling (AQMAS) strategy, enabling a team of agents to intelligently select devices with specific model quantization levels for each round of local training, taking into account the resource constraints of the current devices.
CVPR 2026
CCF-A
Acceptance Ratio: 25%
Computer Vision
NeurIPS 2024
CCF-A
Acceptance Ratio: 25%
Machine Learning
We obtain image high-frequency features through the Discrete Wavelet Transform (DWT) to generate backdoor triggers.
ICCAD 2024
CCF-B
Acceptance Ratio: 24%
Embedded System / EDA
We propose an energy-aware FL framework named DR-FL, which considers the energy constraints in both clients and heterogeneous deep learning models to enable energy-efficient FL. Unlike Vanilla FL, DR-FL adopts our proposed Multi-Agent Reinforcement Learning (MARL)-based dual-selection method...
DAC 2024
CCF-A
Acceptance Ratio: 25%
Embedded System
In this paper, we propose a novel framework to select and store the most representative data online in a self-supervised way. Such data has a small memory footprint and allows infrequent requests of user annotations for further fine-tuning.
IJCAI 2022
CCF-A
Acceptance Ratio: 14%
Artificial Intelligence
We introduce a novel backdoor defense framework named Attention Relation Graph Distillation (ARGD), which fully explores the correlation among attention features with different orders using our proposed Attention Relation Graphs (ARGs).
DAC 2022
CCF-A
Acceptance Ratio: 23%
Embedded System
We fully exploit the parallel processing capability of the underlying hardware to enable a quick search for a barrier certificate.
TCAD 2022
CCF-A
Acceptance Ratio: 23%
Embedded System Journal
We propose a novel framework named PervasiveFL that enables efficient and effective FL among heterogeneous IoT devices. Without modifying original local models, PervasiveFL installs one lightweight NN model named modellet on each device.
ACM MM 2022
CCF-A
Acceptance Ratio: 25%
Multi-Media
In this paper, we propose a novel two-stage backdoor defense method, named MCLDef, based on Model-Contrastive Learning (MCL). MCLDef can purify the backdoored model by pulling the feature representations of poisoned data towards those of their clean data counterparts.
๐ Educations
- 2023.09 - 2026.03 Postdoc, University of Notre Dame, Department of CSE. Advisor: Prof. Yiyu Shi.
- 2019.09 - 2023.06 PhD, East China Normal University, Department of Software Engineering (A-level subject rating). Advisor: Prof. Mingsong Chen.
- 2016.09 - 2019.06 Master, Jiangnan University, Department of Internet of Things, Computer Science. Advisor: Prof. Zhilei Chai.
๐ Honors and Awards
- 2026.07 The OpenAI Researcher Program (PI).
- 2024.12 The NAIRR Pilot Award (PI) and 2026.10 Sustainable Research Pathways (Research Mentor).
- 2023.03 China National Scholarship ($4280)
- 2023.06 Shanghai Outstanding Graduates
- 2022.06 PhD Outstanding Program (ECNU) PI ($4280)
- 2021.06 Outstanding Student (ECNU)
๐ป Research Internships
- 2016.12 - 2019.06 Peking University, Supervisor: Prof. Wei Yan.
๐ฌ Invited Talks & Services
- 2026.11 ICCD 2026 Session Chair.
- 2026.03 Invited Talk at Shandong University.
- 2024.10 ICCAD 2024-2026 Session Chair.
- 2022.12 Invited Talk at East China Normal University.
๐ Teachings
- 2025 Fall Official Instructor, CSE 60685, University of Notre Dame.
- 2025 Spring Official Instructor, CSE 60685, University of Notre Dame.
- 2024 Spring Teaching Assistant, CSE 60685, University of Notre Dame.
๐ Reviews and TPC Members
๐ป Computer Architecture & Hardware
- 2023โ2026 IEEE/ACM International Conference on Computer-Aided Design (ICCAD) TPC Member
- 2026 IEEE International Conference on Computer Design (ICCD) TPC Member
- 2026 IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) TPC Member
- IEEE Transactions on Computers (TC) Reviewer
- High-Confidence Computing Reviewer
๐ค Artificial Intelligence & Machine Learning
- 2026 ICLR Reviewer
- 2026 AAAI TPC Member
- 2023โ2026 Conference on Neural Information Processing Systems (NeurIPS) Reviewer
- 2026 Machine Learning Reviewer
- 2026 Neurocomputing Reviewer
- 2026 Pattern Recognition Reviewer