mlsanityData to 14 Oct 2026

PPFPL: Cross-silo Privacy-preserving Federated Prototype Learning Against Data Poisoning Attacks

arXiv 2504.03173 ↗First posted 4 Apr 2025cs.CR cs.DCcited by 1

On the map: LLM safety and security → Federated unlearning and poisoning

Abstract (arXiv)

Privacy-Preserving Federated Learning (PPFL) enables multiple clients to collaboratively train models by submitting secreted model updates. Nonetheless, PPFL is vulnerable to data poisoning attacks due to its distributed training paradigm in cross-silo scenarios. Existing solutions have struggled to improve the performance of PPFL under poisoned Non-Independent and Identically Distributed (Non-IID) data. To address the issues, this paper proposes a privacy-preserving federated prototype learning framework, named PPFPL, which enhances the cross-silo FL performance against poisoned Non-IID data while protecting client privacy. Specifically, we adopt prototypes as client-submitted model updates to eliminate the impact of poisoned data distributions. In addition, we design a secure aggregation protocol utilizing homomorphic encryption to achieve Byzantine-robust aggregation on two servers, significantly reducing the impact of malicious clients. Theoretical analyses confirm the convergence and privacy of PPFPL. Experimental results on public datasets show that PPFPL effectively resists data poisoning attacks under Non-IID settings.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Federated unlearning and poisoning, one of 12 directions in LLM safety and security. Builds on 22 of its 79 references that are in the map. First posted on arXiv 4 Apr 2025.

Builds on

22 of 79 references.

References · 22

  1. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,349
  2. 2018Federated Optimization in Heterogeneous Networkscited by 2,987
  3. 2020Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimizationcited by 906
  4. 2018Byzantine-Robust Distributed Learning: Towards Optimal Statistical Ratescited by 711
  5. 2021Federated Learning Based on Dynamic Regularizationcited by 562
  6. 2019A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protectioncited by 374
  7. 2021FedProto: Federated Prototype Learning across Heterogeneous Clientscited by 352
  8. 2020Data Poisoning Attacks Against Federated Learning Systemscited by 275
  9. 2018Mitigating Sybils in Federated Learning Poisoningcited by 161
  10. 2020LDP-Fed: Federated Learning with Local Differential Privacycited by 132
  11. 2018Robust Classification with Convolutional Prototype Learningcited by 123
  12. 2022Federated Learning with Label Distribution Skew via Logits Calibrationcited by 109
10 more
  1. 2021FedProc: Prototypical Contrastive Federated Learning on Non-IID datacited by 81
  2. 2020Local Differential Privacy based Federated Learning for Internet of Thingscited by 75
  3. 2021TenSEAL: A Library for Encrypted Tensor Operations Using Homomorphic Encryptioncited by 52
  4. 2021Privacy-preserving Federated Learning based on Multi-key Homomorphic Encryptioncited by 48
  5. 2021Fusion of Federated Learning and Industrial Internet of Things: A Surveycited by 29
  6. 2023Federated Learning without Full Labels: A Surveycited by 13
  7. 2023FLTracer: Accurate Poisoning Attack Provenance in Federated Learningcited by 8
  8. 2023Model Inversion Attack via Dynamic Memory Learningcited by 7
  9. 2023Dynamic Heterogeneous Federated Learning with Multi-Level Prototypescited by 7
  10. 2024FedRFQ: Prototype-Based Federated Learning with Reduced Redundancy, Minimal Failure, and Enhanced Qualitycited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Defending Against Backdoor Attacks via Alignment Checking in Model-Contrastive Federated Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2021PPFL: Privacy-preserving Federated Learning with Trusted Execution Environments
  2. 2024No Vandalism: Privacy-Preserving and Byzantine-Robust Federated Learning
  3. 2023FheFL: Fully Homomorphic Encryption Friendly Privacy-Preserving Federated Learning with Byzantine Users
  4. 2023WW-FL: Secure and Private Large-Scale Federated Learning
  5. 2025SMTFL: Secure Model Training to Untrusted Participants in Federated Learning
  6. 2025Practical Framework for Privacy-Preserving and Byzantine-robust Federated Learning
  7. 2026FedFG: Privacy-Preserving and Robust Federated Learning via Flow-Matching Generation
  8. 2025FuSeFL: Fully Secure and Scalable Federated Learning