
LECTURER · TONGLING UNIVERSITY
Qingkui
Zeng.
Learning together.
Keeping data private.
I study privacy protection and federated learning, with an interest in reliable, efficient and secure learning systems.
ABOUT
Useful learning.
Thoughtful privacy.
I am a Lecturer in the School of Artificial Intelligence at Tongling University. My research focuses on privacy protection and federated learning: how to learn from distributed data while protecting individuals and keeping collaborative systems reliable.
I received my Ph.D. in Information and Communication Engineering from Nanjing University of Information Science and Technology in June 2025, and joined Tongling University in July 2025.
My work spans local differential privacy, poisoning defenses, efficient learning for connected devices, and the security of language model applications.
Google Scholar snapshot · 5 October 2026. See current metrics
RESEARCH
Questions that connect my work.
Privacy, robustness and efficiency
across collaborative learning systems.
Privacy & Differential Privacy
Protecting individual information while retaining useful learning signals, with a focus on local differential privacy and privacy–utility tradeoffs.
Robust Federated Learning
Understanding poisoning threats and designing defenses for collaborative learning across heterogeneous clients.
Learning at the Edge
Exploring efficient, privacy-aware learning for connected devices, healthcare sensing and the Internet of Things.
LLM & Content Security
Studying how instructions hidden in uploaded content can influence language model behavior and compromise application workflows.
NEWS
Recent updates.
Publications and academic milestones.
Our work on efficient federated learning with local differential privacy appeared in Computer Networks. View paper
Our Prompt-in-Content study was published in the NSS 2025 proceedings. View paper
I joined the School of Artificial Intelligence at Tongling University as a Lecturer.
I completed my Ph.D. in Information and Communication Engineering at Nanjing University of Information Science and Technology.
PUBLICATIONS
Work in print.
Journal and conference papers. Years follow the publisher’s final citation; earlier records and conference years are noted where they differ.
Showing all 19 publications
Efficient privacy-preserving federated learning with local differential privacy
Computer Networks, 285, Article 112351, 2026
View citation
@article{zeng202614,
title = {{Efficient privacy-preserving federated learning with local differential privacy}},
author = {Yin, Chunyong and Zeng, Qingkui},
journal = {Computer Networks},
year = {2026},
volume = {285},
eid = {112351},
doi = {10.1016/j.comnet.2026.112351},
url = {https://doi.org/10.1016/j.comnet.2026.112351}
}A Game-Theoretic Framework for Personalized Privacy-Utility Tradeoffs in Federated Learning
Human-centric Computing and Information Sciences, 16, pp. 1–17, 2026
View citation
@article{zeng202619,
title = {{A Game-Theoretic Framework for Personalized Privacy-Utility Tradeoffs in Federated Learning}},
author = {Li, Haiqing and Yin, Chunyong and Zeng, Qingkui and Qi, Chenglong},
journal = {Human-centric Computing and Information Sciences},
year = {2026},
volume = {16},
pages = {1--17},
eid = {41},
doi = {10.22967/HCIS.2026.16.041},
url = {https://doi.org/10.22967/HCIS.2026.16.041}
}Enhancing Privacy and Accountability in Federated Learning for Smart IoT
Proceedings of the Fourth International Conference on Advances in Computing Research (ACR’26), pp. 441–443, 2026
View citation
@inproceedings{zeng202617,
title = {{Enhancing Privacy and Accountability in Federated Learning for Smart IoT}},
author = {Lian, Zhuotao and Zeng, Qingkui and Su, Chunhua},
booktitle = {Proceedings of the Fourth International Conference on Advances in Computing Research (ACR’26)},
year = {2026},
pages = {441--443},
doi = {10.1007/978-3-032-23883-2_36},
url = {https://doi.org/10.1007/978-3-032-23883-2_36}
}Layerwise Selective Quantization for Communication-Efficient Federated Learning in Consumer IoT Devices
2026 IEEE International Conference on Consumer Electronics (ICCE), pp. 1–5, 2026
View citation
@inproceedings{zeng202618,
title = {{Layerwise Selective Quantization for Communication-Efficient Federated Learning in Consumer IoT Devices}},
author = {Lian, Zhuotao and Zeng, Qingkui and Su, Chunhua},
booktitle = {2026 IEEE International Conference on Consumer Electronics (ICCE)},
year = {2026},
pages = {1--5},
doi = {10.1109/icce67443.2026.11449814},
url = {https://doi.org/10.1109/icce67443.2026.11449814}
}POSTER: A Server-Side Proactive Defense Framework for Poison-Resilient Federated Learning
Provable and Practical Security (ProvSec 2025), pp. 498–502, 2026
First online 2025; final citation 2026
View citation
@inproceedings{zeng202616,
title = {{POSTER: A Server-Side Proactive Defense Framework for Poison-Resilient Federated Learning}},
author = {Zeng, Qingkui and Lian, Zhuotao},
booktitle = {Provable and Practical Security (ProvSec 2025)},
year = {2026},
pages = {498--502},
doi = {10.1007/978-981-95-2961-2_29},
url = {https://doi.org/10.1007/978-981-95-2961-2_29}
}Prompt-in-Content Attacks: Exploiting Uploaded Inputs to Hijack LLM Behavior
Network and System Security (NSS 2025), pp. 146–160, 2026
NSS 2025 proceedings; published February 2026
View citation
@inproceedings{zeng202611,
title = {{Prompt-in-Content Attacks: Exploiting Uploaded Inputs to Hijack LLM Behavior}},
author = {Lian, Zhuotao and Wang, Weiyu and Zeng, Qingkui and Nakanishi, Toru and Kitasuka, Teruaki and Su, Chunhua},
booktitle = {Network and System Security (NSS 2025)},
year = {2026},
pages = {146--160},
doi = {10.1007/978-981-95-6419-4_9},
url = {https://doi.org/10.1007/978-981-95-6419-4_9}
}FLNAP: Mitigating LDP Impact in Federated Learning with Noise-Aware Pre-Training
2025 17th International Conference on Advanced Infocomm Technology (ICAIT), pp. 128–134, 2025
View citation
@inproceedings{zeng202515,
title = {{FLNAP: Mitigating LDP Impact in Federated Learning with Noise-Aware Pre-Training}},
author = {Zeng, Qingkui and Lian, Zhuotao and Niu, Huijuan and Song, Jingcheng and Zhao, Hong and Su, Chunhua},
booktitle = {2025 17th International Conference on Advanced Infocomm Technology (ICAIT)},
year = {2025},
pages = {128--134},
doi = {10.1109/icait66450.2025.11353293},
url = {https://doi.org/10.1109/icait66450.2025.11353293}
}Privacy-Enhanced Federated WiFi Sensing for Health Monitoring in Internet of Things
IEEE Internet of Things Journal, 12(3), pp. 2994–3002, 2025
Scholar record 2024; final citation 2025
View citation
@article{zeng202512,
title = {{Privacy-Enhanced Federated WiFi Sensing for Health Monitoring in Internet of Things}},
author = {Lian, Zhuotao and Zeng, Qingkui and Liu, Zhusen and Wang, Haoda and Ma, Chuan and Meng, Weizhi and Su, Chunhua and Sakurai, Kouichi},
journal = {IEEE Internet of Things Journal},
year = {2025},
volume = {12},
number = {3},
pages = {2994--3002},
doi = {10.1109/jiot.2024.3476149},
url = {https://doi.org/10.1109/jiot.2024.3476149}
}Defending Against Data Poisoning Attack in Federated Learning With Non-IID Data
IEEE Transactions on Computational Social Systems, 11(2), pp. 2313–2325, 2024
Scholar record 2023; final citation 2024
View citation
@article{zeng202404,
title = {{Defending Against Data Poisoning Attack in Federated Learning With Non-IID Data}},
author = {Yin, Chunyong and Zeng, Qingkui},
journal = {IEEE Transactions on Computational Social Systems},
year = {2024},
volume = {11},
number = {2},
pages = {2313--2325},
doi = {10.1109/tcss.2023.3296885},
url = {https://doi.org/10.1109/tcss.2023.3296885}
}Traffic Sign Recognition Using Optimized Federated Learning in Internet of Vehicles
IEEE Internet of Things Journal, 11(4), pp. 6722–6729, 2024
Scholar record 2023; final citation 2024
View citation
@article{zeng202405,
title = {{Traffic Sign Recognition Using Optimized Federated Learning in Internet of Vehicles}},
author = {Lian, Zhuotao and Zeng, Qingkui and Wang, Weizheng and Xu, Dequan and Meng, Weizhi and Su, Chunhua},
journal = {IEEE Internet of Things Journal},
year = {2024},
volume = {11},
number = {4},
pages = {6722--6729},
doi = {10.1109/jiot.2023.3312348},
url = {https://doi.org/10.1109/jiot.2023.3312348}
}Blockchain-Based Two-Stage Federated Learning With Non-IID Data in IoMT System
IEEE Transactions on Computational Social Systems, 10(4), pp. 1701–1710, 2023
Scholar record 2022; final citation 2023
View citation
@article{zeng202302,
title = {{Blockchain-Based Two-Stage Federated Learning With Non-IID Data in IoMT System}},
author = {Lian, Zhuotao and Zeng, Qingkui and Wang, Weizheng and Gadekallu, Thippa Reddy and Su, Chunhua},
journal = {IEEE Transactions on Computational Social Systems},
year = {2023},
volume = {10},
number = {4},
pages = {1701--1710},
doi = {10.1109/tcss.2022.3216802},
url = {https://doi.org/10.1109/tcss.2022.3216802}
}Hybrid Representation and Decision Fusion towards Visual-textual Sentiment
ACM Transactions on Intelligent Systems and Technology, 14(3), pp. 1–17, 2023
View citation
@article{zeng202309,
title = {{Hybrid Representation and Decision Fusion towards Visual-textual Sentiment}},
author = {Yin, Chunyong and Zhang, Sun and Zeng, Qingkui},
journal = {ACM Transactions on Intelligent Systems and Technology},
year = {2023},
volume = {14},
number = {3},
pages = {1--17},
doi = {10.1145/3583076},
url = {https://doi.org/10.1145/3583076}
}DEEP-FEL: Decentralized, Efficient and Privacy-Enhanced Federated Edge Learning for Healthcare Cyber Physical Systems
IEEE Transactions on Network Science and Engineering, 9(5), pp. 3558–3569, 2022
View citation
@article{zeng202201,
title = {{DEEP-FEL: Decentralized, Efficient and Privacy-Enhanced Federated Edge Learning for Healthcare Cyber Physical Systems}},
author = {Lian, Zhuotao and Yang, Qinglin and Wang, Weizheng and Zeng, Qingkui and Alazab, Mamoun and Zhao, Hong and Su, Chunhua},
journal = {IEEE Transactions on Network Science and Engineering},
year = {2022},
volume = {9},
number = {5},
pages = {3558--3569},
doi = {10.1109/tnse.2022.3175945},
url = {https://doi.org/10.1109/tnse.2022.3175945}
}Hybrid Anomaly Detection via Multihead Dynamic Graph Attention Networks for Multivariate Time Series
IEEE Access, 10, pp. 40967–40978, 2022
View citation
@article{zeng202203,
title = {{Hybrid Anomaly Detection via Multihead Dynamic Graph Attention Networks for Multivariate Time Series}},
author = {Zhou, Liwen and Zeng, Qingkui and Li, Bo},
journal = {IEEE Access},
year = {2022},
volume = {10},
pages = {40967--40978},
doi = {10.1109/access.2022.3167640},
url = {https://doi.org/10.1109/access.2022.3167640}
}Privacy-Enhanced Federated Generative Adversarial Networks for Internet of Things
The Computer Journal, 65(11), pp. 2860–2869, 2022
View citation
@article{zeng202208,
title = {{Privacy-Enhanced Federated Generative Adversarial Networks for Internet of Things}},
author = {Zeng, Qingkui and Zhou, Liwen and Lian, Zhuotao and Huang, Huakun and Kim, Jung Yoon},
journal = {The Computer Journal},
year = {2022},
volume = {65},
number = {11},
pages = {2860--2869},
doi = {10.1093/comjnl/bxac060},
url = {https://doi.org/10.1093/comjnl/bxac060}
}Privacy-preserving Blockchain-based Global Data Sharing for Federated Learning with Non-IID Data
2022 IEEE 42nd International Conference on Distributed Computing Systems Workshops (ICDCSW), pp. 193–198, 2022
View citation
@inproceedings{zeng202207,
title = {{Privacy-preserving Blockchain-based Global Data Sharing for Federated Learning with Non-IID Data}},
author = {Lian, Zhuotao and Zeng, Qingkui and Su, Chunhua},
booktitle = {2022 IEEE 42nd International Conference on Distributed Computing Systems Workshops (ICDCSW)},
year = {2022},
pages = {193--198},
doi = {10.1109/icdcsw56584.2022.00044},
url = {https://doi.org/10.1109/icdcsw56584.2022.00044}
}WebFed: Cross-platform Federated Learning Framework Based on Web Browser with Local Differential Privacy
ICC 2022 - IEEE International Conference on Communications, pp. 2071–2076, 2022
View citation
@inproceedings{zeng202206,
title = {{WebFed: Cross-platform Federated Learning Framework Based on Web Browser with Local Differential Privacy}},
author = {Lian, Zhuotao and Yang, Qinglin and Zeng, Qingkui and Su, Chunhua},
booktitle = {ICC 2022 - IEEE International Conference on Communications},
year = {2022},
pages = {2071--2076},
doi = {10.1109/icc45855.2022.9838421},
url = {https://doi.org/10.1109/icc45855.2022.9838421}
}Local Differential Privacy for Data Streams
Security and Privacy in Digital Economy, pp. 143–160, 2020
View citation
@inproceedings{zeng202010,
title = {{Local Differential Privacy for Data Streams}},
author = {Fang, Xianjin and Zeng, Qingkui and Yang, Gaoming},
booktitle = {Security and Privacy in Digital Economy},
year = {2020},
pages = {143--160},
doi = {10.1007/978-981-15-9129-7_11},
url = {https://doi.org/10.1007/978-981-15-9129-7_11}
}Local differential privacy for human-centered computing
EURASIP Journal on Wireless Communications and Networking, 2020(1), Article 65, 2020
View citation
@article{zeng202013,
title = {{Local differential privacy for human-centered computing}},
author = {Fang, Xianjin and Zeng, Qingkui and Yang, Gaoming},
journal = {EURASIP Journal on Wireless Communications and Networking},
year = {2020},
volume = {2020},
number = {1},
eid = {65},
doi = {10.1186/s13638-020-01675-8},
url = {https://doi.org/10.1186/s13638-020-01675-8}
}No matching publications.
Try a broader keyword or reset the filters.
RESEARCH HIGHLIGHTS
Three threads. One shared goal.
Making collaborative and intelligent
systems more dependable.
PRIVACY-PRESERVING LEARNING
FDLDPComputer Networks · 2026
A better starting point for private learning.
A two-stage framework uses locally private reports to initialize a global model, then protects client updates during federated fine-tuning. The work studies the connection between privacy, convergence and communication.
Read the paperROBUST FEDERATED LEARNING
FedCleanerProvSec 2025 · 2026
Detect, then remove harmful influence.
A server-side defense combines proactive layer-wise anomaly detection with retroactive contribution erasure to address poisoned updates and their influence on the global model.
Read the paperLLM SECURITY
Prompt-in-ContentNSS 2025 · 2026
When uploaded content becomes an instruction.
This study examines adversarial prompts embedded in otherwise ordinary inputs, how they redirect model outputs, and why content isolation matters for safer LLM applications.
Read the paperCURRICULUM VITAE
Education & experience.
01 / EDUCATION
Sep 2021 – Jun 2025
Ph.D. in Information and Communication Engineering
Nanjing University of Information Science and Technology
02 / EMPLOYMENT
Jul 2025 – Present
Lecturer
Tongling University
School of Artificial Intelligence
BEYOND RESEARCH
A little more about me.
Music, mountain trails
and the colors of porcelain.

SOUND
Electric Guitar & Heavy Music
I enjoy playing electric guitar, especially rock, metal and modern heavy music. Riffs, rhythm and the search for a good tone are a welcome change of pace.

OUTDOORS
Hiking & Time Outdoors
Hiking gives me time to step away from the screen, follow a trail and take in the details of the natural world. I enjoy the quieter pace of time outdoors.

MATERIALS
Ceramic Glazes & Kiln Firing
I am interested in porcelain glaze colors, surface textures and kiln firing. I enjoy exploring how materials and heat shape the final appearance of a ceramic piece.
CONTACT
Let’s exchange ideas.
I welcome conversations about privacy protection, federated learning and secure learning systems.