arXiv Computation and Language

Component-Aware Differential Privacy for Federated Multilingual Speech-LLMs

The paper introduces a per-layer differential privacy (DP) clipping strategy for federated multilingual speech large language models (speech‑LLMs). It demonstrates that standard single‑pool per‑layer DP methods fail due to a cross‑component budget collapse caused by large norm differences between acoustic encoders and language decoders. The authors propose an α‑split two‑pool allocation that normalises encoder and decoder parameters separately, preserving the overall DP guarantee while restoring word error rate performance and providing tighter noise protection for the encoder.

arXiv Machine Learning

2d ago

Is Weight Tying Still Beneficial for Decoder-Only LLMs in Private Settings Under DP-SGD?

arXiv:2609.40335v1 Announce Type: new Abstract: Differentially Private Stochastic Gradient Descent (DP-SGD) is a leading approach for privacy-preserving fine-tuning of large language models (LLMs). M...

By Razan El Mais, Ali Chehab, Ibrahim Issa, Razane Tajeddine

llms rag fine-tuning safety

arXiv Machine Learning

Aug 5

Noise-Aware Shrinkage for Differentially Private Zeroth-Order Fine-Tuning of Large Language Models

arXiv:2608. 03277v1 Announce Type: new Abstract: Differentially private zeroth-order optimization (DP-ZO) enables memory-efficient private fine-tuning of large language models using only forward evaluations.

By Lele Zheng, Weifeng Kong, Xinyi Zhang, Ke Cheng, Tao Zhang, Yulong Shen

llms fine-tuning safety

arXiv Machine Learning

Jul 31

TriShield: Zero-Utility-Loss Defense Against Privacy Backdoors in Federated Language Model Fine-Tuning via Orthogonal Gradient Projection and Optimizer State Entanglement

arXiv:2607. 27940v1 Announce Type: new Abstract: Federated fine-tuning of large language models (LLMs) enables collaborative training without exposing raw data.

By Cheng Wei (Honor Device Co., Ltd., Shenzhen, China)

llms fine-tuning safety

Hugging Face Trending Papers

Jul 30

TriShield: Zero-Utility-Loss Defense Against Privacy Backdoors in Federated Language Model Fine-Tuning via Orthogonal Gradient Projection and Optimizer State Entanglement

Federated fine-tuning of large language models (LLMs) enables collaborative training without exposing raw data. However, a recent attack, NeuroImprint [1] (arXiv:2606.

llms fine-tuning safety

arXiv Machine Learning

Aug 18

An Adaptive Gradient Clipping and Noise Injection Mechanism for Differentially Private Federated Learning

arXiv:2608. 15153v1 Announce Type: cross Abstract: Differentially private federated learning must balance privacy protection against model accuracy and training efficiency.

By Wenjing Wei, Alla Jammine, Farid Nait-Abdesselam

safety

arXiv Machine Learning

Jul 23

End-to-End Differential Privacy in Training Deep Neural Network Classifiers

arXiv:2607. 19580v1 Announce Type: new Abstract: Differentially private machine learning enables model training on sensitive data while ensuring that individual data is unlikely to be recoverable from the parameters of the resulting model.

By Huaiyuan Rao, Calvin Hawkins, Alexander Benvenuti, Matthew Hale

benchmarks safety

Hugging Face Trending Papers

Jun 1

IntraShuffler: A Privacy Preserving Framework for Heterogeneous DP Federated Learning

Heterogeneous Differential Privacy (HDP) in Federated Learning (FL) allows clients to select individual privacy budgets ($\varepsilon_i$) according to institutional policies and data sensitivity. In practice, many HDP-FL systems employ $\varepsilon$-aware server aggregation to improve model utility by re-weighting client updates according to their declared privacy budgets.

safety

arXiv Machine Learning

Jul 23

Differentially Private Neural Network Training Under the Hidden State Assumption

arXiv:2407. 08233v3 Announce Type: replace Abstract: Current differentially private learning paradigms face a severe utility bottleneck: DP-SGD degrades performance through noise accumulation over training steps, while aggregation-based approaches such as PATE suffer from data inefficiency due to disjoint data partitioning.

By Ding Chen, Haochen Luo, Xiaofei Wang, Chen Liu

benchmarks safety

arXiv Machine Learning

Sep 15

Privacy-Preserving Deep Joint Source-Channel Coding with In-Loop Concept Erasure

arXiv:2609.13393v1 Announce Type: cross Abstract: Deep joint source-channel coding (DeepJSCC) transmits learned semantic features efficiently but can leak sensitive attributes such as gender, race, o...

By Rami Eid, Maria Slim, Mariette Awad, Hadi Sarieddeen

safety

arXiv Machine Learning

Sep 15

Machine Unlearning for Speech Question Answering in Large Audio-Language Models

arXiv:2609.13195v1 Announce Type: new Abstract: Large Audio-Language Models (LALMs) have recently shown strong capabilities in speech understanding and question answering (QA), but they also inherit...

By Zhe Liu

llms nlp fine-tuning multimodal benchmarks safety

arXiv Machine Learning

Sep 10

RAPTOR: Role-Aware Private Training for Mixture-of-Experts

arXiv:2609.05770v1 Announce Type: new Abstract: Differentially private (DP) fine-tuning methods treat sparse Mixture-of-Experts (MoE) models as a single dense block, ignoring that shared layers see a...

By Duc Dm, Khai Le-Duc, Nguyen Do, Minh Son Hoang, Florent Draye, Thai Hoang, Hoang Phuong Dam, Jiarui Liu, Chris Ngo, Terry Jingchen Zhang, Anh Le Duc Tran, Nhat Do Minh, Minh Ngoc Le, My T. Thai, Ran Xu, Silvio Savarese, Mona Diab, Bernhard Sch\"olkopf, Zhijing Jin, Huy L. Nguyen, Daeyoung Kim

llms fine-tuning safety

arXiv Machine Learning

Aug 3

StraightDP: Geometry-Aware Differential Privacy for Rectified-Flow Transformers

arXiv:2607. 29100v1 Announce Type: new Abstract: Differentially private (DP) training of text-conditioned generative models suffers a utility cliff at strong privacy.

By Xujun Che, Depeng Xu, Xintao Wu

llms diffusion fine-tuning multimodal safety

Related stories