The DOI matches its Crossref record; 44 of the 47 references it cites were found in a registry or an index.
Privacy-Enhanced Federated Learning Against Poisoning Adversaries
Abstract
Federated learning (FL), as a distributed machine learning setting, has received considerable attention in recent years. To alleviate privacy concerns, FL essentially promises that multiple parties jointly train the model by exchanging gradients rather than raw data. However, intrinsic privacy issue still exists in FL, e.g., user’s training samples could be revealed by solely inferring gradients. Moreover, the emerging poisoning attack also poses a crucial security threat to FL. In particular, due to the distributed nature of FL, malicious users may submit crafted gradients during the training process to undermine the integrity and availability of the model. Furthermore, there exists a contradiction in simultaneously addressing two issues, that is, privacy-preserving FL solutions are dedicated to ensuring gradients indistinguishability, whereas the defenses against poisoning attacks tend to remove outliers based on their similarity. To solve such a dilemma, in this paper, we aim to build a bridge between the two issues. Specifically, we present a privacy-enhanced FL (PEFL) framework that adopts homomorphic encryption as the underlying technology and provides the server with a channel to punish poisoners via the effective gradient data extraction of the logarithmic function. To the best of our knowledge, the PEFL is the first effort to efficiently detect the poisoning behaviors in FL under ciphertext. Detailed theoretical analyses illustrate the security and convergence properties of the scheme. Moreover, the experiments conducted on real-world datasets show that the PEFL can effectively defend against label-flipping and backdoor attacks, two representative poisoning attacks in FL.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“To solve this problem, [38], [49], [51] used homomorphic encryption to evaluate the local model securely.”
From Poisoning attacks in federated learning: A survey (2023) · cites this paper as [51]
Citations
Cites
Advances and Open Problems in Federated Learning
Kairouz, P., & McMahan, H. B. (2021) · Foundations and Trends in Machine Learning · doi:10.1561/2200000083 · Cited by 7,772
Public-Key Cryptosystems Based on Composite Degree Residuosity Classes
Paillier, P. (2007) · Lecture Notes in Computer Science · doi:10.1007/3-540-48910-x_16 · Cited by 7,250
Deep Learning with Differential Privacy
Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., & Zhang, L. (2016) · Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2976749.2978318 · Cited by 6,578
ADADELTA: An Adaptive Learning Rate Method
Zeiler, M. D. (2012) · arXiv · doi:10.48550/arxiv.1212.5701 · Cited by 5,322
Practical Secure Aggregation for Privacy-Preserving Machine Learning
Bonawitz, K., Ivanov, V., Kreuter, B., Marcedone, A., McMahan, H. B., Patel, S., Ramage, D., Segal, A., & Seth, K. (2017) · Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3133956.3133982 · Cited by 3,956
Pearson Correlation Coefficient
Benesty, J., Chen, J., Huang, Y. A., & Cohen, I. (2009) · Springer topics in signal processing · doi:10.1007/978-3-642-00296-0_5 · Cited by 3,540
Privacy-preserving deep learning
Shokri, R., & Shmatikov, V. (2015) · Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2810103.2813687 · Cited by 2,356
On the momentum term in gradient descent learning algorithms
Ning Qian, A. (1999) · Neural Networks · doi:10.1016/s0893-6080(98)00116-6 · Cited by 2,318
SecureML: A System for Scalable Privacy-Preserving Machine Learning
Mohassel, P., & Zhang, Y. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.12 · Cited by 1,936
Nasr, M., Shokri, R., & Houmansadr, A. (2019) · 2019 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2019.00065 · Cited by 1,760
Cited by
Sorted by citations
Items 201–300 of 351
Li, X., Tao, X., Chen, G., You, L., & Sun, Y. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3573910 · Cited by 4
“Instead of eliminating malicious updates, Liu et al. [27] leveraged coordinate-wise median as a benchmark, calculate the similarity between the benchmark and each uploaded gradient, and correspondingly, adjust the aggregation weight to stabilize the learning.”
InfluentialFederated Client-Tailored Adapter for Medical Image Segmentation
Hu, G., Song, S., Kang, Y., Yin, Z., Zhao, G., Li, C., & Tang, J. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3581432 · Cited by 3
“Federated Learning (FL) is one typical decentralized training technique, which collectively learns a global model in a central server from multiple distributed clients without sharing the distributed local data [12], [13], [14].”
Enhancing Security in Parallel Federated Learning With Sharded Blockchain for Internet of Vehicles
Tian, L., Lin, F., Gan, J., Qi, J., Jia, R., Zheng, Z., Li, M., & Shen, X. (2025) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2025.3593493 · Cited by 3
“Liu et al. [9] introduced PEFL, leveraging Homomorphic Encryption (HE) and adaptive aggregation to resist model poisoning.”
Badar, H. M. S., Kajla, N. I., Mahmood, K., Khan, M. W., Fan, G., & Zhang, C. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3644701 · Cited by 4
“Compromised participants can craft updates that corrupt aggregation and degrade clinical reliability; recent works report enhanced and adaptive attacks tailored to federated protocols [3], [4], [5], [6].”
Peng, J., Xing, H., Xu, L., Luo, S., Dai, P., Feng, L., Ale, L., Zhao, B., & Xiao, Z. (2026) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2026.3666663 · Cited by 3
“Liu et al. [59] presented PEFL, a privacy-enhanced FL framework adopting homomorphic encryption and providing the server with a channel to punish poisoners by removing the malicious gradients.”
Federated learning with bilateral defense via blockchain
Xiao, J., Nie, H., Yi, Z., Tang, X., & Lu, S. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.107199 · Cited by 4
A distributed learning framework with blockchain and privacy-preserving for IoV
Li, C., Long, Y., Ding, Y., Yang, C., Zhang, C., Shen, M., & Zhu, L. (2025) · Applied Soft Computing · doi:10.1016/j.asoc.2025.113710 · Cited by 3
LZKSA: Lattice-Based Special Zero-Knowledge Proofs for Secure Aggregation's Input Verification
Lu, Z. G., & Lu, S. (2025) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3719027.3744812 · Cited by 2
Luo, S., Fu, S., Luo, Y., Liu, L., Deng, Y.-M., & Wang, S. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-39828-5_11 · Cited by 3
PPFL-IDS: Privacy-Preserving Federated Learning Based IDS Against Poisoning Attacks
Xu, M., & Li, X. (2023) · Mobile Networks and Applications · doi:10.1007/s11036-023-02231-6 · Cited by 3
“AlthoughsomePPFL studies canresist poisoning attacks [28–30], there is no research on the corresponding poisoning attack and anti-attack strategy for intrusion detection systems.”
Order Optimal Bounds for One-Shot Federated Learning Over Non-Convex Loss Functions
Sharifnassab, A., Salehkaleybar, S., & Golestani, S. J. (2023) · IEEE Transactions on Information Theory · doi:10.1109/tit.2023.3344141 · Cited by 3
“Moreover, some notions of privacy can be preserved by utilizing differential privacy techniques [26], [27], [28], [29], [30], [31], [32], [33], [34], [35] or secure multi-party computation [36], [37], [38], [39].”
BackgroundFedValidate: A Robust Federated Learning Framework Based on Client-Side Validation
Zhu, W., Liu, Z., Chen, Z., Shi, C., Zhang, X., & Guo, S. (2023) · IEEE International Conference on Data Science in Cyberspace · doi:10.1109/dsc59305.2023.00055 · Cited by 3
“Liu et al. [29] proposed PEFL, which requires more hardware resources and involves two servers, adding extra communication overhead.”
BackgroundSelf-citationFedChallenger: Challenge-Response-Based Defence for Federated Learning Against Byzantine Attacks
Moyeen, M. A., Kaur, K., Agarwal, A., Ricardo Manzano, S., Zaman, M., & Goel, N. (2023) · Globecom · doi:10.1109/globecom54140.2023.10437878 · Cited by 3
“Liu et al. [10] discard model updates by detecting the difference between benign and malicious updates using the Pearson correlation coefficient.”
MethodOptimizing federated learning with weighted aggregation in aerial and space networks
Dong, F., Leung, H., & Drew, S. (2024) · Journal of Network and Computer Applications · doi:10.1016/j.jnca.2024.104086 · Cited by 3
Zhao, R., Yang, L. T., Liu, D., et al. (2024) · IEEE Network · doi:10.1109/mnet.2024.3521004 · Cited by 2
“For example, in MNIST dataset with label-flipping attack, the feature spaces of a source label 1 could be flipped to the target label 9 for misclassification [6].”
Hu, Y., Wang, Z., Xia, Y., Xiong, H., Fang, L., Peng, C., Wahaballa, A., Qin, Z., & Qin, Z. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3579613 · Cited by 2
“In addition, FL can be combined with secure multiparty computation (MPC) [10], [11], homomorphic encryption (HE) [12], [13], and differential privacy (DP) [14], [15] to enhance data and model confidentiality while providing robust defense against inference attacks.”
Xue, K., Li, J., Xue, R., Xue, Y., & Zhao, J. (2025) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2025.3632902 · Cited by 3
“Liu et al. [7] assigned aggregation weights via median and Pearson correlation metrics.”
DamPa: Dynamic Adaptive Model Poisoning Attack in Federated Learning
Hu, F., Zhang, A., Liu, X., & Li, M. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3631449 · Cited by 2
“In model poisoning attacks [7], [8], adversaries submitted malicious local updates, steering the convergence of the aggregated gradient to cause arbitrary deviations from the correct trends [27].”
Self-citationPPFPL: Cross-Silo Privacy-Preserving Federated Prototype Learning Against Data Poisoning Attacks
Zhang, H., Yu, J., Xu, F., Hu, C., Zhang, Y., Wang, X., Yu, Z., & Zhang, X. (2025) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2025.3643391 · Cited by 2
“For instance, the work in [15] proposed a privacy-enhanced FL framework that adopts HE as the underlying technology and provides two servers to punish malicious clients via gradient extraction of logarithmic function.”
InfluentialFRIDA: Free-rider detection using privacy attacks
Recasens, P. G., Horváth, Á., Gutierrez-Torre, A., Torres, J., Berral, J. L., & Pejó, B. (2025) · Journal of Information Security and Applications · doi:10.1016/j.jisa.2025.104357 · Cited by 3
Zhao, W., Pang, Q., Sun, Y., Li, X., Li, B., & Zhang, Z. (2026) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2026.3651570 · Cited by 2
“Traditional bit-level communication approaches have explored homomorphic encryption (HE) to secure gradient information against GIAs [9], [10].”
Two-Dimensional Privacy-Preserving Federated Learning Scheme Against Poisoning Attacks
Gu, K., Zhao, W., Tan, J., Yang, Z., Li, X., & Weijia Jia, B. (2026) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2026.3656703 · Cited by 3
“Liu et al. [20] employed homomorphic encryption technology and solved the contradictory relationship between protecting client privacy and defending against poisoning attacks.”
Li, X., Li, X., & Feng, X. (2026) · Alexandria Engineering Journal · doi:10.1016/j.aej.2026.02.004 · Cited by 2
Secure and Privacy-Preserving Federated Learning for Next-Generation Underground Mine Safety
Elmahallawy, M., Madria, S., & Frimpong, S. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11402420 · Cited by 2
“Homomorphic Encryption (HE) [29] HE allows computations to be performed on encrypted data without needing to decrypt it.”
PoFEL: Energy-Efficient Consensus for Blockchain-Based Hierarchical Federated Learning
Li, S., Hu, Q., Wang, Z., Xu, M., & Cai, Z. (2026) · IEEE Transactions on Services Computing · doi:10.1109/tsc.2026.3675091 · Cited by 3
“However, the implementation of HFL still faces two major challenges: i) the single cloud server risks the whole system with the single-point failure; and ii) the layered topology introduces additional model security and privacy issues, such as model poisoning attacks by altering the intermediate models [5] and leaking…”
LSFL: A Lightweight and Secure Federated Learning scheme for Internet of Vehicles
Wei, L., Peng, D., Shi, Y., Shi, L., Li, G. Y., & Lian, H. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2026.104858 · Cited by 2
Wang, M., Cao, C., Wang, X., Zhang, Q., Jing, Z., Li, H., & Sun, J. (2023) · Mathematics · doi:10.3390/math11214547 · Cited by 2
Self-citationMS-FL: A Federated Learning Framework Based on Multiple Security Strategies
Yang, W.-T., Kang, P., & Wei, C. (2024) · IEEE Access · doi:10.1109/access.2024.3353131 · Cited by 3
“Therefore, in the rest of this section, Algorithm 6 Model Updating we provide hybrid argument and simulator [16] to further demonstrate that during the execution of MS-FL protocols, model requestor and service provider cannot learn nothing about data owners’ privacy.”
MethodAFL: Attention-Based Byzantine-Robust Federated Learning with Vector Filter
Chen, H., Lv, X., & Zheng, W. (2023) · Globecom · doi:10.1109/globecom54140.2023.10437587 · Cited by 2
“According to [15], we let the distance between w ∗ and w τ be h , i.e., h = w ∗ − w τ .”
Privacy-Preserving and Reliable Distributed Federated Learning
Dong, Y., Zhang, L., & Xu, L. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-0834-5_9 · Cited by 2
An Enclave-Aided Byzantine-Robust Federated Aggregation Framework
Yao, J., Song, C., Li, H., Wang, Y., Yang, Q., & Wang, L. (2024) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc57260.2024.10570631 · Cited by 2
“For example, [13] adopts HE as the underlying technology and integrates basic robust aggregation.”
BackgroundMethodInfluentialSelf-citationA Differential Privacy Federated Learning Scheme with Improved Noise Perturbation
Liu, C., He, X., Wang, B., et al. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-5606-3_6 · Cited by 2
Privacy Preservation in User Behavior Analysis for Mobile-Edge Computing
Deng, S., Zhang, J., & Zhang, L. (2024) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2024.3472608 · Cited by 2
“Definition 6: According to [46], we let X be the original gradient, and Y be the safety gradient, then PCCs = Cov ( X , Y ) is said to be the Pearson correlation coefficient between the original gradient and the safety gradient.”
SecDefender: Detecting low-quality models in multidomain federated learning systems
Sameera, K. M., Sgueglia, A., Vinod, P., A., R. R. K., Visaggio, C. A., Sorbo, A. D., & Conti, M. (2024) · Future Generation Computer Systems · doi:10.1016/j.future.2024.107587 · Cited by 2
Sinaei, S., Mohammadi, M., Shrestha, R., Alibeigi, M., & Eklund, D. (2024) · 2024 27th Euromicro Conference on Digital System Design (DSD) · doi:10.1109/dsd64264.2024.00063 · Cited by 2
“The encrypted results are then returned to devices for decryption and aggregation [13].”
MethodOvercoming Data Imbalance in Federated Learning with Calibration Weighting
Kang, N., Lim, Y., & Im, J. (2024) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata62323.2024.10825694 · Cited by 2
Guo, J., Yang, M., Li, H., Wu, X., He, Y., & Wang, X. (2024) · International Conference on Computer Science and Management Technology (ICCSMT) · doi:10.1145/3708036.3708276 · Cited by 2
“Liu et al. [13] proposed the PEFL framework, which assesses gradient similarity using the Pearson correlation coefficient and removes malicious gradients.”
Self-citationAccess Microservices with Zero-Knowledge, Path Certainty and Distributed Traceability
Quach, V., Dantu, R., Talapuru, S., Zaman, S., & Pokharel, A. (2025) · IEEE International Symposium on Service-Oriented System Engineering (SOSE) · doi:10.1109/sose67019.2025.00015 · Cited by 2
“Privacy-Enhanced Federated Learning (PEFL) framework [11] uses Homomorphic Encryption (HE) to protect privacy while enabling poisoning attack detection.”
Zhang, H., Yu, Z., Wang, G., Xu, F., Zhang, Y., Hu, C., Wang, X., & Yu, J. (2025) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2025.3620529 · Cited by 2
“Incontrast, HE-based FL solutions [21] [42] adopt HE for privacy protection, and their privacy computation overhead can be expressed as (cid:16) O ( T ng ) .”
2021 Index IEEE Transactions on Information Forensics and Security Vol. 16
(2021) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2022.3143280 · Cited by 2
Research on spatial-temporal data fusion algorithm of sensors
Li, T. (2024) · International Conference on High Performance Computing and Communication Engineering (HPCCE) · doi:10.1117/12.3026624 · Cited by 1
“From a mathematical point of view, the values of statistical data are arranged in chronological order, with emphasis on the sequence of data [4] .”
BackgroundA Review on Privacy Enhanced Distributed ML Against Poisoning Attacks
Archa, A. T., & Kartheeban, K. (2024) · Lecture notes in networks and systems · doi:10.1007/978-981-97-3973-8_11 · Cited by 1
A Verifiable Federated Learning Algorithm Supporting Distributed Pseudonym Tracking
Xie, H., Wang, Y., Ding, Y., et al. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-5562-2_11 · Cited by 1
Defending Against Poisoning Attacks in Federated Prototype Learning on Non-IID Data
Zhang, J., Zhang, H., Wang, G., & Dong, A. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-71467-2_6 · Cited by 1
A Survey on Securing Image-Centric Edge Intelligence
Li, T., Hu, H., Gabbouj, M., Ye, Q., Xiang, Y., Li, J., & Li, L. (2024) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3700792 · Cited by 1
Integrating CP-ABE and Device Fingerprint Into Federated Learning
Chen, C., Román, R., Wang, K. I., & Sakurai, K. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-2417-1_23 · Cited by 1
Xie, Q., & Zhao, Y. (2025) · Computer Networks · doi:10.1016/j.comnet.2025.111493 · Cited by 2
Zhao, Y., Cao, Y., Zhang, J., Zhang, Z., & Wang, W. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129502 · Cited by 1
Resisting against targeted poisoning attacks in lightweight privacy preserving federated learning
Zhang, H., Xie, H., & Lv, J. (2025) · The Journal of Supercomputing · doi:10.1007/s11227-025-07829-z · Cited by 2
“In addition, FL usually suffers from statistical heterogeneity of data distributions (i.e., Non-IID), resulting in inconsistent model updates among clients.”
InfluentialDeception-Based Defense of Poisoning Attack in Federated Learning
Masse, G. C. T., Benslimane, A., Tchendji, V. K., & Hemida, A. H. A. (2025) · IEEE Access · doi:10.1109/access.2025.3611806 · Cited by 1
“Methods such as clipping extreme gradients and applying differential privacy are significant passive defenses [16], [26].”
A Comparative Analysis of Label Flipping and Backdoor Data Poisoning Attacks in Federated Learning
Bisna, N. D., Navaneeth, D., & James, A. (2025) · International Conference on Soft Computing for Security Applications (ICSCSA) · doi:10.1109/icscsa66339.2025.11171343 · Cited by 1
“2) Privacy-Enhanced Federated Learning (PEFL): Liu et al. [7] propose PEFL, which preserves privacy while enabling poisoning detection through encrypted updates.”
SignDefence: Byzantine-Robust Federated Learning with Sign Direction and Leaky ReLU
Moyeen, M. A., Kaur, K., Agarwal, A., Ricardo Manzano, S., Zaman, M., & Goel, N. (2025) · International Conference on Communications · doi:10.1109/icc52391.2025.11161809 · Cited by 1
“The method of Liu et al. [13] uses Pearson’s correlation to distinguish between benign and malicious updates.”
DRIFT: DCT-based robust and intelligent federated learning with trusted privacy
Dong, Q., Bai, Y., Su, M., Gao, Y., & Fu, A. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.131697 · Cited by 1
Shree, A. N. R., Macharla, S. R., Monteiro, D. H., Yathin, K. N., Vishwanath, H., & Biradar, A. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-3306-9.ch009 · Cited by 1
Federated Learning in Modern Education: Balancing Privacy, Scalability, and Effectiveness
Ganatra, R. G., Gahlawat, M., & Dudhagara, C. R. (2025) · Studies in systems, decision and control · doi:10.1007/978-3-031-99270-4_11 · Cited by 1
Masse, G. C. T., Benslimane, A., Tchendji, V. K., Hemida, A. H. A., Su, Z., & Han, S. (2025) · International Conference on Communications · doi:10.1109/icc52391.2025.11161880 · Cited by 1
“This is achieved through model poisoning, where they introduce a lot of harmful local model updates during the training process to disrupt the global model optimization algorithm [12].”
Peng, C., & Xian, H. (2025) · Journal of Information Security and Applications · doi:10.1016/j.jisa.2025.104345 · Cited by 2
PDFL: A Privacy-Enhancing and Robust Poisoning Defense Federated Learning Scheme
Wang, G., Wu, H., Li, Q., Liu, Z., Zhao, J., Lam, K., Dong, Q., Yang, Y., & Zhou, L. J. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3648147 · Cited by 1
“Reference [21] defenses against poisoning attacks and tends to remove outliers based on Pearson correlation coe ffi - cient, which can e ff ectively defend against label-flipping and backdoor attacks.”
SHIFT: Enhancing federated learning robustness through client-side backdoor detection
Wang, K., Wang, L., Liu, Z., Luo, Y., Zhang, K., & Li, W. (2026) · Information Fusion · doi:10.1016/j.inffus.2026.104144 · Cited by 1
FedD2: Data Poisoning Robust Defense Strategy in Federated Learning
Wu, J. F., Liu, Y., Ni, T., Zheng, X., Luo, Y., & Hu, P. (2025) · IEEE International Conference on Parallel and Distributed Systems (ICPADS) · doi:10.1109/icpads67057.2025.11323206 · Cited by 1
SA-FL: Secure Aggregation Scheme in Federated Learning Against Poisoning Attacks
Chang, Y., Chen, Q., Luan, T. H., & Su, Z. (2026) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2026.3656284 · Cited by 1
“Recent advances reveal that even when gradients are privatized using simple multiplicative blinding, statistical correlations, such as the Pearson correlation coe ffi cient (PCC), can be largely preserved [14], [15].”
Differentially Private Federated Learning with Adaptive Noise Scheduling
J, S. J., Kaushik, S., Mishra, D., & Bhowmick, S. (2025) · IEEE Pune Section International Conference (PuneCon) · doi:10.1109/punecon67554.2025.11378972 · Cited by 1
“However, despite its advantages, it can still disclose sensitive information to the central server, thus introducing data privacy issues, such as inference attacks [8] or data poisoning attack [9].”
TCFL: Trapdoor homomorphic encryption based federated learning aggregator with filter
Shen, X., Jia, Y., Xu, C. L., & Zhu, L. (2026) · Peer-to-Peer Networking and Applications · doi:10.1007/s12083-025-02193-4 · Cited by 1
Jia, X., Sun, R. X., Liu, L., Liu, J., Kang, J., Mumtaz, S., & Dong, M. (2026) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2026.3671265 · Cited by 1
“Other schemes, such as PEFL [31], FLOD [32], AMMFRS [33] adopt various privacy-preserving strategies, but they suffer from similar drawbacks: performance degradation, high overhead, or limited applicability due to rigid assumptions.”
Chen, L., Batool, H., Zeng, X., Zhao, D., Du, Y., & Tan, C. W. (2026) · Applied Soft Computing · doi:10.1016/j.asoc.2026.114982 · Cited by 1
BloBS-FL: A Robust Architecture of Blockchain Based Decentralized Split Federated Learning
Mishra, S., Prajapati, Y., Aggarwal, K., et al. (2026) · doi:10.1109/comsnets67989.2026.11418208 · Cited by 1
“We simulate adversarial behavior where malicious clients perform label-flipping attacks [23] to access robustness.”
PRFed: Privacy-preserving federated learning to resist poisoning attack
Li, H., Chen, Y., Yang, Y., Dou, H., & Yang, Y. (2026) · Information Fusion · doi:10.1016/j.inffus.2026.104397 · Cited by 1
Self-citationSu, Z., Lu, Z. G., Wu, Y. H., Shen, R., & Lu, S. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00373 · Cited by 1
“PEFL [ 11].”
From data heterogeneity to convergence: A data-centric review of federated learning
Nguyen, H. T., Bettinelli, M., Ghaffari, A., Benoît, A., Nguyen, T., Pirttikangas, S., & Lovén, L. (2026) · Neurocomputing · doi:10.1016/j.neucom.2026.135005 · Cited by 1
A Byzantine-Robust Secure Federated Learning Scheme in Heterogeneous Data
Wang, R., Wang, R., Li, Z., Zhang, F., Li, J., & Li, X. (2026) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2026.3714122 · Cited by 1
“2) Defense Baselines: We use 7 Byzantine-robust aggregation algorithms as comparison baselines, namely FedAvg [1], Krum [18], Median [17], T-mean [17], ShieldFL [10], PEFL [9], and RFLPA [26].”
InfluentialFedMC: Federated Learning with Mode Connectivity Against Distributed Backdoor Attacks
Wang, W., Zhang, C., Liu, S., Tang, M., Liu, A., & Yu, S. (2023) · International Conference on Communications · doi:10.1109/icc45041.2023.10278903 · Cited by 1
“Homomorphic encryption and corresponding detection methods [8, 9] were proposed to punish the malicious clients’ models based on the encrypted gradients.”
MethodEfficient and Privacy-Preserving Byzantine-Robust Federated Learning
Luan, S., Lu, X., Zhang, Z., Chang, G., & Guo, Y. (2023) · Globecom · doi:10.1109/globecom54140.2023.10437426 · Cited by 1
“For example, PEFL proposed by [11] performs Median [15] aggregation rules based on HE protocols to defend against Byzantine attacks without any information leakage.”
MethodSun, H., Wang, Y., Yang, H., et al. (2024) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2024.3389611 · Cited by 1
“In homomorphic encryption field, PEFL [17] is the first effort to efficiently detect the poisoning behaviors against malicious adversaries.”
BackgroundDye4AI: Assuring Data Boundary on Generative AI Services
Wang, S., Sun, K., & Zhai, Y. (2024) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3658644.3670299 · Cited by 1
“Federated learning (FL) only transmits model parameters, allowing decentralized model training without the need to centralize raw data in a single location [52, 58].”
Supvirus: A Scenario-Oriented Feature Poisoning Attack Approach in SplitFed Learning
Zhao, Y., Wu, G., Hou, J., & Li, X.-Y. (2023) · International Conference on Mobility, Sensing and Networking (MSN) · doi:10.1109/msn60784.2023.00073 · Cited by 1
“1) Untargeted Data Poisoning Attack: Untargeted data poisoning attack refers to the attacker interfering with the model training accuracy by introducing malicious data samples into training data before the training phase [12], [18]–[22].”
BackgroundZhong, Y., & Wang, L.-P. (2024) · International Conference on Computer Supported Cooperative Work in Design (CSCWD) · doi:10.1109/cscwd61410.2024.10580526 · Cited by 1
“For instance, PEFL [7] utilizes the Pearson correlation coefficient to identify outliers, employing a dual-server model for secure computation where trusted servers hold keys to collaborate with additive homomorphic encryption algorithms.”
MethodData Privacy Preserving for Centralized Robotic Fault Diagnosis With Modified Dataset Distillation
Wang, T., Huang, Y., Liu, Y., & Chen, C. (2024) · Journal of Computing and Information Science in Engineering · doi:10.1115/1.4066096 · Cited by 1
HIPFL: A Hierarchical Incentive-Based Personalized Federated Learning
Liu, Y., Bao, Z., Zhang, W., Yao, S., Zhao, H., Zhang, Z., & Miao, Y. (2024) · IEEE International Conference on Digital Twins and Parallel Intelligence (DTPI) · doi:10.1109/dtpi61353.2024.10778797 · Cited by 1
Privacy-Preserving and Robust Federated Learning Based on Secret Sharing
Mei, J., Shen, X., Xu, C. L., Zhu, L., Jin, G., & Sharif, K. (2024) · IEEE International Symposium on Parallel and Distributed Processing with Applications (ISPA) · doi:10.1109/ispa63168.2024.00223 · Cited by 1
“Liu et al. [16] proposed the PEFL scheme, using homomorphic encryption as the underlying technology to protect clients’ privacy.”
Research Directions for Decentralized Technology Transactions: An Update
Hadi, A., Rahmadika, S., Rahmi, U., Larasati, H. T., & Fajri, B. R. (2025) · Salud Ciencia y Tecnología · doi:10.56294/saludcyt20252161 · Cited by 1
A Federated Learning Backdoor Attack Defense Method Based on Dual Attention Mechanism
Yan, J., Yingchi, M., Nie, H., Tu, Z., & Huang, J. (2022) · IEEE International Conference on Big Data Computing Service and Applications (BigDataService) · doi:10.1109/bigdataservice55688.2022.00030 · Cited by 1
Xu, M. (2022) · International Conference on Networking and Network Applications (NaNA) · doi:10.1109/nana56854.2022.00089 · Cited by 1
“In order to protect the privacy of the normal user gradient while reducing the impact of malicious poison on the global model, existing work first determines a benchmark of benign gradient, and then calculates similarity of different local upload gradient and this gradient, excludes the gradient of low similarity, and…”
BackgroundDetecting Low-Quality Models in Multidomain Federated Learning SystemsPreprint
Sameera, K. M., Sgueglia, A., Vinod, P., Rehiman, K. A. R., Visaggio, C. A., & Sorbo, A. D. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4604411
Privacy-preserving Federated Learning Against Byzantine Attack via Top-k Indexes
Shang, M., Tong, Z., Liu, T., & Zhang, T. (2023) · International Conference on Networking and Network Applications (NaNA) · doi:10.1109/nana60121.2023.00108
“PEFL [21] uses the dual-server model for secure computation using homomorphic encryption.”
MethodOptimizing Federated Learning Approach: Literature Survey and Open Points
El-Sayed, T., Moustafa, A., El-Rashidy, M. A., & El‐Sayed, A. (2023) · International Conference on Electronic Engineering (ICEEM) · doi:10.1109/iceem58740.2023.10319542
“[37] created an outlier data point detection technique that efficiently eliminates tag reversal and backdoor-based poisoning attempts.”
Efficient Homomorphic Convolution for Secure Deep Learning Inference
Liu, X., Li, H., Qian, Q., et al. (2023) · International Conference on Privacy, Security and Trust (PST) · doi:10.1109/pst58708.2023.10320171
“Fruitful advancements have been made in this field, including [4], [9], [14], [16], [20], [22], [23], [25], [32].”
BackgroundSelf-citationDeep Federated Learning Based on Knowledge Distillation and Differential Privacy
Lin, H., Yu, F., & Wang, X. (2023) · doi:10.1002/9781119913924.ch5
Marulli, F., Campanile, L., Marrone, S., & Verde, L. (2024) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-3-031-57931-8_29
Refine and Identify: An Accelerated Iterative Algorithm for Securing Federated Learning
Gouissem, A., Chkirbene, Z., Khattab, T. M., Mabrok, M. A., Abdallah, M. M., & Hamila, R. (2024) · International Wireless Communications and Mobile Computing Conference (IWCMC) · doi:10.1109/iwcmc61514.2024.10592366
“Encryption mechanisms, as proposed in [14] and [15], introduce an additional layer of security by leveraging homomorphic encryption and secure aggregation protocols to shield the FL process from poisoning attacks and ensure user privacy.”
BackgroundA Survey of Federated Learning: Review, Attacks, Defenses
Yao, Z., Cheng, J., Fu, C., & Huang, Z. (2024) · Communications in computer and information science · doi:10.1007/978-981-97-4387-2_13
Lightweight Byzantine-Robust and Privacy-Preserving Federated Learning
Lu, Z. G., Lu, S., Cui, Y., Wu, J., Nie, H., Xiao, J., & Yi, Z. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-69766-1_19
Invisible Backdoor Attack against 3D Point Cloud by Optimized 3D Transformation
Ao, Z., & Yufeng, L. (2024) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn60899.2024.10651278
“Similar to attaching patches in the corners of images in 2D backdoor attacks [28], they place additional points around 3D objects like spheres to inject into 3D deep learning [32], [33].”
BackgroundDecentralized Edge-Based Detection of Label Flipping Attacks in Federated Learning
Alotaibi, N., Felemban, M., & Mahmood, S. (2024) · Lecture notes in networks and systems · doi:10.1007/978-3-031-75016-8_19
FEFL: A Fair & Efficient Defense Against Poisoning Attacks via Discrete Update Space
Zhao, Y., Zhang, J., Cao, Y., & Ye, T. (2024) · International Conference on Information Communication and Signal Processing (ICICSP) · doi:10.1109/icicsp62589.2024.10809244
“…some existing defense methods mainly embark on the following four aspects: 1) direct identification of client attributes (malicious or benign) [5,6]; 2) detection of anomalous data or model parameters [7,8]; 3) design of robust aggregation rules [9,10]; 4) privacy-oriented perspective [11,12].”
Privacy-Preserving Byzantine-Robust Federated Learning via Multiparty Homomorphic Encryption
Jiang, W., Luo, S., Fu, S., Liu, L., & Luo, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-1093-8_36
Qi, M., Wang, X., He, Y., Yang, M., & Wu, X. (2024) · IEEE International Conference on Frontiers Technology of Information and Computer (ICFTIC) · doi:10.1109/icftic64248.2024.10913255
“Privacy protection in FL primarily relies on three methods: Differential Privacy (DP) [7]–[9], Secure Multiparty Computation (SMC) [10], and Homomorphic Encryption (HE) [11], [12].”
Research on Federated Learning Scheme Based on Secure Aggregation
Zhang, X., Song, H., & Wan, F. (2024) · IEEE International Conference on Frontiers Technology of Information and Computer (ICFTIC) · doi:10.1109/icftic64248.2024.10912977
“At present, researchers have proposed a variety of secure aggregation schemes based on cryptographic techniques, such as secret sharing [6], homomorphic encryption [7,8], secure multi-party computation [9], mask-based and other methods.”
Federated Learning in der Arbeitsplanung
Hussong, M., Klar, M., & Aurich, J. C. (2025) · Zeitschrift für wirtschaftlichen Fabrikbetrieb · doi:10.1515/zwf-2024-0131
SecureFed: Blockchain-Based Defence for Data Poisoning Attack in Federated Learning
Alsharidah, A. A., Jha, D. N., Wei, B. W., Solaiman, E., & Ranjan, R. K. (2024) · IEEE/ACM International Conference on Utility and Cloud Computing (UCC) · doi:10.1109/ucc63386.2024.00064
“Robust aggregation algorithms [5], [6], [7], Anomaly detection techniques [8], [9], [10], [11] and Blockchain-based solutions [12], [13], [14] are the most commonly used.”
LBRFL: Lightweight Privacy-Preserving Federated Learning with Byzantine-Robustness
Shen, P., Zhao, S., Xu, J., Zhao, C., Chen, Z., & Guo, S. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-6606-5_3
Self-citation
Metadata
- Authors
- Liu, X., Li, H., Xu, G., Chen, Z., Huang, X., & Lu, R.
- Affiliations
- University of Electronic Science and Technology of China (CN)
- Nanyang Technological University (SG)
- China Electronics Technology Group Corporation (CN)
- University of New Brunswick (CA)
- Venue
- IEEE Transactions on Information Forensics and Security
- Published
- 2021
- Volume
- 16
- Pages
- 4574-4588
- ISSN
- 1556-6013 (ISSN-L)
- 1556-6021
- OpenAlex
- W3198262235
- Cited by
- 351
- References
- 47
- Citations by year
- 2026: 47 · 2025: 102 · 2024: 110 · 2023: 66 · 2022: 23 · 2021: 3
- FWCI
- 17.85
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Information Forensics and Security (publishedVersion)
- DR-NTU (Nanyang Technological University) (submittedVersion)
- Funders
- National Natural Science Foundation of China (62020106013, 61972454, 61802051, 61772121, 61728102)
- Sichuan Science and Technology Program (2020JDTD0007, 2020YFG0298)
- Chongqing Science and Technology Commission (cstc2018jcyjAX0703)
- Fundamental Research Funds for Chinese Central Universities (ZYGX2020ZB027)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Physical Sciences › Computer Science › Computer Networks and Communications › Cryptography and Data Security
- Keywords
- privacy-preserving federated learning, poisoning attacks, homomorphic encryption, gradient privacy, poisoning detection, label-flipping attacks, backdoor attacks
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
