Bingyan Liu
	Associate Professor (Tenured), PhD Advisor Beijing University of Posts and Telecommunications (BUPT)     Address: School of Computer Science, BUPT (Xituchen campus), Haidian District, Beijing, China, 100876 Email: bingyanliu AT bupt.edu.cn

BRIEF BIOGRAPHY

Currently I am an associate professor at School of Computer Science, Beijing University of Posts and Telecommunications. I received my PhD degree in Computer Science from Peking University in 2022, advised by Prof. Yao Guo and Prof. Xiangqun Chen. I previously interned in MSRA working with Senior Researcher Yuanchun Li and Principle Researcher Yunxin Liu. My research focuses on systems infrastructure for ubiquitous and trustworthy AI, aiming to build systems that enable large models to operate as low-latency, energy-efficient, privacy-preserving control loops on mobile phones, robots, and edge devices. My current research has three directions:

• Agent & Embodied AI Systems: on-device agents, mobile LLM runtimes, VLA/WAM serving, real-time robot control.
• Trustworthy Edge AI: TEE-assisted ML, secure on-device inference, privacy-preserving personalization.
• Federated and Distributed AI: robust, traceable, personalized, and device-aware federated/distributed learning.

SELECTED PUBLICATIONS

• [ICML'26] Reflex: Real-Time Vision-Language-Action Control through Streaming Inference. (CCF-A, full paper)
  Yuanchun Guo, Bingyan Liu.
  Forty-Third International Conference on Machine Learning, 2026.
• [ICASSP'26] FedSKU: Defending backdoors in federated learning through selective knowledge unlearning. (CCF-B, full paper)
  Guofu Xie, Yu Zhou, Bingyan Liu.
  ICASSP 2026-2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026.
• [ICASSP'26] DeMoFL: Efficient and Effective Decentralized Model-Heterogeneous Federated Learning. (CCF-B, full paper)
  Yuanchun Guo, Bingyan Liu.
  ICASSP 2026-2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026.
• [TMC'26] PFHAR: Practically Adopting Multi-Modal Foundation Model for Human Activity Recognition through Edge-cloud Collaborative Learning. (CCF-A, full paper)
  Zhengyuan Zhang, Dong Zhao, Guanzhou Zhu, Xiangyu Li, Chunliang Li, Bingyan Liu, Yuanchun Li, Huadong Ma.
  IEEE Transactions on Mobile Computing.
• [ICMR'25] MoAFCL: Feature-Aware Mixture-of-Adapter for Federated Continual Learning. (CCF-B, full paper)
  Dian Zhang, Bingyan Liu.
  Proceedings of the 2025 International Conference on Multimedia Retrieval, 2025.
• [ICASSP'25] Multiple sclerosis detection with reinforcement learning and differential evolution. (CCF-B, full paper)
  Jing Yang, Jin Yang, Chenwei Wu, Gaozhe Jiang, Yaning Lv, Bingyan Liu, Feng Xu.
  In Proceedings of ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2025.
• [DAC'25] PracMHBench: Re-evaluating Model-Heterogeneous Federated Learning Based on Practical Edge Device Constraints. (CCF-A, full paper)
  Yuanchun Guo, Bingyan Liu, Yulong Sha, Zhensheng Xian.
  In Proceedings of Design Automation Conference (DAC), June. 2025.
• [AAAI'25] PA3Fed: Period-Aware Adaptive Aggregation for Improved Federated Learning. (CCF-A, full paper)
  Chengxiang Huang, Bingyan Liu.
  In Proceedings of AAAI Conference on Artificial Intelligence (AAAI), Feb. 2025.
• [KDD'25] BTFL: A Bayesian-based Test-Time Generalization Method for Internal and External Data Distributions in Federated learning. (CCF-A, full paper)
  Yu Zhou, Bingyan Liu.
  In Proceedings of KDD, Aug. 2025.
• [Neurocomputing'24] Recent advances on federated learning: A systematic survey. (JCR-Q1, full paper)
  Bingyan Liu, Nuoyan Lv, Yuanchun Guo, and Yawen Li.
  In Neurocomputing, June. 2024.
• [BIB'24] Image-based Molecular Representation Learning for Drug Development: A Survey. (JCR-Q1/CCF-B, full paper)
  Yue Li, Bingyan Liu, Jinyan Deng, Yi Guo, and Hongbo Du.
  In Briefing In Bioinformatics, May. 2024.
• [Security'24] FAMOS:Robust Privacy-Preserving Authentication on Payment Apps via Federated Multi-Modal Contrastive Learning. (CCF-A, full paper)
  Yifeng Cai, Ziqi Zhang, Jiaping Gui, Bingyan Liu, Xiaoke Zhao, Ruoyu Li, Zhe Li, and Ding Li.
  In Proceedings of USENIX Security Symposium (Security), May. 2024.
• [CVPR'24] Traceable Federated Continual Learning. (CCF-A, full paper)
  Qiang Wang, Bingyan Liu, and Yawen Li.
  In IEEE / CVF Computer Vision and Pattern Recognition Conference, Feb. 2024.
• [IEEE S&P'24] No Privacy Left Outside: On the (In-)Security of TEE-Shielded DNN Partition for On-Device ML. (CCF-A, full paper)
  Ziqi Zhang, Chen Gong, Yifeng Cai, Yuan yuanyuan, Bingyan Liu, Shuai Wang, Ding Li, Yao Guo, and Xiangqun Chen.
  In IEEE Symposium on Security and Privacy, July. 2023.
• [TKDE'23] Multi-view Scholar Clustering with Dynamic Interest Tracking. (CCF-A, full paper)
  Ang Li, Yawen Li, Yingxia Shao, and Bingyan Liu.
  In IEEE Transactions on Knowledge and Data Engineering, Feb. 2023.
• [WWW'23] Beyond Fine-Tuning: Efficient and Effective Fed-Tuning for Mobile/Web Users. (CCF-A, full paper)
  Bingyan Liu, Yifeng Cai, Hongzhe Bi, Ziqi Zhang, Ding Li, Yao Guo, and Xiangqun Chen.
  In Proceedings of the 32th Web Conference, Jan. 2023.
• [ICSE'23] FedSlice: Protecting Federated Learning Models from Malicious Participants with Model Slicing. (CCF-A, full paper)
  Ziqi Zhang, Yuanchun Li, Bingyan Liu, Yifeng Cai, Ding Li, Yao Guo, and Xiangqun Chen.
  In International Conference on Software Engineering, Dec. 2022.
• [ISSTA'22] TEESlice: Slicing DNN Models for Secure and Efficient Deployment. (CCF-A, full paper)
  Ziqi Zhang, Lucien K. L. Ng, Bingyan Liu, Yifeng Cai, Ding Li, Yao Guo, and Xiangqun Chen.
  In International Symposium on Software Testing and Analysis(workshop), June. 2022.
• [ICSE'22] ReMoS: Reducing Defect Inheritance in Transfer Learning via Relevant Model Slicing. (CCF-A, full paper)
  Ziqi Zhang, Yuanchun Li, Jindong Wang, Bingyan Liu, Ding Li, Yao Guo, Xiangqun Chen, and Yunxin Liu.
  In International Conference on Software Engineering, Dec. 2021.
• [ACM Ubicomp'22] DistFL: Distribution-aware Federated Learning for Mobile Scenarios. [PDF] (CCF-A, full paper)
  Bingyan Liu, Yifeng Cai, Ziqi Zhang, Yuanchun Li, Leye Wang, Ding Li, Yao Guo, and Xiangqun Chen.
  In ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, Oct 2021.
• [ISSTA'21] ModelDiff: Testing-based DNN Similarity Comparison for Model Reuse Detection. [PDF] (CCF-A, full paper, 51/233≈21.9%)
  Yuanchun Li, Ziqi Zhang, Bingyan Liu, Ziyue Yang, Yunxin Liu.
  In Proceeding of the 30th ACM SIG-SOFT International Symposium on Software Testing and Analysis.
• [WWW'21] PFA: Privacy-preserving Federated Adaptation for Effective Model Personalization. [PDF] (CCF-A, full paper, 357/1736≈20.6%)
  Bingyan Liu, Yao Guo, and Xiangqun Chen.
  In Proceedings of the 30th Web Conference.
• [ACM Ubicomp'21] PMC: A Privacy- preserving Deep Learning Model Customization Framework for Edge Computing. [PDF] (CCF-A, full paper, 149/848≈17.5%)
  Bingyan Liu, Yuanchun Li, Yunxin Liu, Yao Guo, and Xiangqun Chen.
  In Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 4, 4, Article 139 (December 2020), 25 pages.
• [AAAI'21] TransTailor: Pruning the Pre-trained Model for Improved Transfer Learning. [PDF] (CCF-A, full paper, 1692/7961≈21.3%)
  Bingyan Liu, Yifeng Cai, Yao Guo, and Xiangqun Chen.
  In Proceedings of the 35th AAAI Conference on Artificial Intelligence.
• [ACM MM'19] WealthAdapt: A General Net-Adaptation Framework for Small Data Tasks. [PDF] (CCF-A, full paper, 248/936≈26.5%)
  Bingyan Liu, Yao Guo, Xiangqun Chen.
  In Proceedings of the 27th ACM International Conference on Multimedia.

INVITED TALKS

• Ant Group, Beijing. Security and Privacy in Split Learning, May 2023
• Huawei, Beijing. Security and Privacy in Federated Learning, Apr 2023
• The 32th Web Conference (WWW), Apr 2023
• ACM Conference on Pervasive and Ubiquitous Computing (UbiComp), Sep 2021
• Introduction of our work on federated personalization, AI Drive, June 2021
• Introduction of our work on model adaptation, AIIT, Hangzhou, May 2021
• The 30th Web Conference (WWW), Virtual Event, Apr 2021
• The 35th AAAI Conference on Artificial Intelligence (AAAI), Virtual Event, Feb 2021
• The 27th ACM International Conference on Multimedia (ACMMM), France, Oct 2019

ACADEMIC SERVICES

• ICML 2025 (CCF-A), Reviewer
• ICLR 2025 (Top-tie), Reviewer
• AAAI 2025 (CCF-A), PC member
• Neurips 2024 (CCF-A), Reviewer
• CVPR 2024,2025 (CCF-A), Reviewer
• KDD 2023,2024,2025 (CCF-A), Reviewer
• WWW 2022,2023,2024,2025 (CCF-A), Reviewer
• Ubicomp 2020,2024 (CCF-A), Reviewer
• ACM MM 2023,2024,2025 (CCF-A), Reviewer

• Nature Communications 2025, Reviewer
• TMC 2024 (CCF-A), Reviewer
• TPAMI 2024 (CCF-A), Reviewer
• TKDE 2022,2023 (CCF-A), Reviewer
• IJCV 2023 (CCF-A), Reviewer
• TOIS 2022 (CCF-A), Reviewer
• JSAC 2022 (CCF-A), Reviewer

TEACHING EXPERIENCE

• Course Instructor, Machine Learning Practical Training, Beijing University of Posts and Telecommunications (Fall 2024)
• Course Instructor, Operating System, Beijing University of Posts and Telecommunications (Fall 2023,Fall 2024)
• Course Instructor, Operating System Practice, Beijing University of Posts and Telecommunications (Spring 2024)
• Teaching Assistant, Operating System, Peking University (Spring 2018 - Fall 2020)

SELECTED HONORS

• Excellent Advisor for Undergraduate Thesis of Beijing, 2024
• Excellent Advisor for Undergraduate Thesis of BUPT, 2024
• Outstanding Graduate of Beijing, 2022
• Outstanding Graduate of Peking University, 2022
• Academic Innovation Award, Peking University, 2021
• PhD National Scholarship, Ministry of Education, 2021
• Merit student, Peking University, 2021
• Huawei Scholarship, Huawei, 2021
• Stars of Tomorrow Internship Program, Microsoft Research Asia, 2020
• Scientific Research Excellence Award, Peking University, 2019
• Scientific Research Excellence Award, Peking University, 2018
• Outstanding Undergraduate Students of Beijing, 2017