# Taming Noise-Induced Prototype Degradation for Privacy-Preserving Personalized Federated Fine-Tuning

arXiv: https://arxiv.org/abs/2604.27833 · HTML: https://mlsanity.com/paper/2604.27833/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 3 of its 56 references that are in the map.
- First posted on arXiv 30 Apr 2026.

## Builds on (most cited first, 3 of 3)

- [Differentially Private Federated Learning: A Client Level Perspective](https://mlsanity.com/paper/1712.07557/) (2017)
- [Effects of Degradations on Deep Neural Network Architectures](https://mlsanity.com/paper/1807.10108/) (2018)
- [FedP3E: Privacy-Preserving Prototype Exchange for Non-IID IoT Malware Detection in Cross-Silo Federated Learning](https://mlsanity.com/paper/2507.07258/) (2025)

## Built on by (most cited first, 1 of 1)

- [Towards Privacy-Preserving Federated Prompt Tuning under Data Heterogeneity: A Subspace-Decomposed Expert Approach](https://mlsanity.com/paper/2607.21417/) (2026)

## Abstract (arXiv)

Prototype-based Personalized Federated Learning (ProtoPFL) enables efficient multi-domain adaptation by communicating compact class prototypes, but directly sharing them poses privacy risks. A common defense involves per-example ℓ2 clipping before prototype computation to bound sensitivity, followed by isotropic Gaussian noise to enforce Local Differential Privacy (LDP). However, Isotropic Gaussian Prototype Perturbation (IGPP) typically over-perturbs discriminative dimensions and struggles to balance the clipping threshold with representation fidelity. In this paper, we propose VPDR, a client-side privacy plug-in that seamlessly integrates into existing ProtoPFLs. Motivated by the observation that dimension-wise class variance reflects discriminability, we introduce Variance-adaptive Prototype Perturbation (VPP), which allocates less noise to discriminative subspaces, preserving semantic separability while ensuring privacy. We further develop Distillation-guided Clipping Regularization (DCR), which enables feature norms to adaptively concentrate near the predefined clipping threshold while maintaining prediction consistency. Theoretical analysis shows that our groupwise mechanism provides privacy guarantees no weaker than the isotropic baseline under the same privacy constraints. Extensive experiments on multi-domain benchmarks demonstrate that VPDR achieves a superior privacy-utility trade-off, outperforming IGPP in personalized federated fine-tuning without sacrificing robustness against realistic attacks.