Pith. sign in

REVIEW 5 cited by

DP-DyLoRA: Fine-Tuning Transformer-Based Models On-Device under Differentially Private Federated Learning using Dynamic Low-Rank Adaptation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2405.06368 v4 pith:ILHGCELO submitted 2024-05-10 cs.LG cs.CRcs.DC

DP-DyLoRA: Fine-Tuning Transformer-Based Models On-Device under Differentially Private Federated Learning using Dynamic Low-Rank Adaptation

classification cs.LG cs.CRcs.DC
keywords privacyfederatedlearningclientsdifferentialfine-tuningadaptationcontributions
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Federated learning (FL) allows clients to collaboratively train a global model without sharing their local data with a server. However, clients' contributions to the server can still leak sensitive information. Differential privacy (DP) addresses such leakage by providing formal privacy guarantees, with mechanisms that add randomness to the clients' contributions. The randomness makes it infeasible to train large transformer-based models, common in modern federated learning systems. In this work, we empirically evaluate the practicality of fine-tuning large scale on-device transformer-based models with differential privacy in a federated learning system. We conduct comprehensive experiments on various system properties for tasks spanning a multitude of domains: speech recognition, computer vision (CV) and natural language understanding (NLU). Our results show that full fine-tuning under differentially private federated learning (DP-FL) generally leads to huge performance degradation which can be alleviated by reducing the dimensionality of contributions through parameter-efficient fine-tuning (PEFT). Our benchmarks of existing DP-PEFT methods show that DP-Low-Rank Adaptation (DP-LoRA) consistently outperforms other methods. An even more promising approach, DyLoRA, which makes the low rank variable, when naively combined with FL would straightforwardly break differential privacy. We therefore propose an adaptation method that can be combined with differential privacy and call it DP-DyLoRA. Finally, we are able to reduce the accuracy degradation and word error rate (WER) increase due to DP to less than 2% and 7% respectively with 1 million clients and a stringent privacy budget of $\epsilon=2$.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Toward Efficient Membership Inference Attacks against Federated Large Language Models: A Projection Residual Approach

    cs.LG 2026-04 unverdicted novelty 7.0

    ProjRes achieves near-100% accuracy in membership inference on FedLLMs by measuring projection residuals of hidden embeddings on gradient subspaces, outperforming prior methods by up to 75.75% even under differential privacy.

  2. Improving Parameter-Efficient Federated Learning with Differentially Private Refactorization

    cs.CR 2026-05 unverdicted novelty 6.0

    FedPower improves the accuracy-privacy tradeoff in differentially private LoRA-based federated learning by reconstructing and clipping full-rank updates then using PowerDP to inject noise before orthonormalization in ...

  3. Differentially Private Clustered Federated Learning with Privacy-Preserving Initialization and Normality-Driven Aggregation

    cs.LG 2026-04 unverdicted novelty 6.0

    PINA improves accuracy in differentially private clustered federated learning by an average of 2.9% using privacy-preserving LoRA sketches for cluster initialization and normality-driven aggregation.

  4. DP-LAC: Lightweight Adaptive Clipping for Differentially Private Federated Fine-tuning of Language Models

    cs.LG 2026-05 unverdicted novelty 5.0

    DP-LAC provides a new adaptive clipping technique for DP-SGD in federated LLM fine-tuning that improves accuracy by 6.6% on average without consuming additional privacy budget or requiring new hyperparameters.

  5. FedShield-LLM: A Secure and Scalable Federated Fine-Tuned Large Language Model

    cs.CR 2025-06 unverdicted novelty 5.0

    FedShield-LLM integrates pruning and FHE on LoRA parameters to support secure, scalable federated fine-tuning of LLMs such as Llama-2.