Standard Differential-Privacy

Follow this topic Ask about this topic

Momentum

1 paper in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 25

All topics

Component-Aware Differential Privacy for Federated Multilingual Speech-LLMs
Sep 11, 2026Jordi Luque, Fernando López, Aleix SantStandard Differential-Privacy Gradient-Based Attacks
Scientific Research Telef´onica Innovaci´on Digital Barcelona, Spain · Telef´onica Innovaci´on Digital Universidad Aut´onoma de Madrid Madrid, Spain · Scientific Research Telef´onica Innovaci´on Digital Universitat Polit`ecnica de Catalunya Barcelona, Spain
Performative Privacy: When Differential Privacy Maximizes Utility
Aug 28, 2026Uddalak Mukherjee, Edwige Cyffers, Yann ChevaleyreStandard Differential-Privacy Privacy-Preserving Machine Learning
Dauphine PSL · CNRS Dauphine PSL
Dithered Gaussian Mechanism for Randomness-Efficient Differential Privacy
Jul 7, 2026Nikita P. Kalinin, Rasmus PaghStandard Differential-Privacy Random
Institute of Science and Technology Austria · BARC University of Copenhagen
Differentially Private Natural Gradient Descent
Jul 7, 2026Pan Li, Kai Chen, Shuai Chang +3Standard Differential-Privacy Stochastic Gradient Descent
Institute of Information Engineering, Chinese Academy of Sciences Beijing, China · Department of Computer Science, Metropolitan College, Boston University Boston, United States · Nanyang Technological University Singapore, Singapore
The Binary Tree Mechanism is Optimal for Approximate Differentially Private Continual Counting
Jul 1, 2026Konstantina Bairaktari, Kasper Green LarsenStandard Differential-Privacy O(\Bar{K}\Log N)$
Department of Computer Science, Aarhus University.
Differential Privacy of Gaussian Process Posterior Sampling
Jun 16, 2026Tomasz MaciazekStandard Differential-Privacy Gaussian Process
School of Mathematics, University of Bristol
Cross-Silo De-Anonymization Under Local Differential Privacy: Threat Model, Phase Transition, and Coordination Necessity
Jun 15, 2026Ziniu Liu, Aiping LiStandard Differential-Privacy Anonymization
National University of Defense Technology
On Choosing the $μ$ Parameter in Gaussian Differential Privacy
Jun 8, 2026Bogdan Kulynych, Antti HonkelaStandard Differential-Privacy Privacy
Biomedical Data Science Center, Lausanne University Hospital · University of Helsinki
DPDL: Towards Differential Privacy Preservation in Decentralized Stochastic Learning on Non-IID Data
Jun 3, 2026Yunsheng Yuan, Xue Xiao, Lina Wang +1Decentralized Learning Standard Differential-Privacy
School of Computer Science and Technology, Shandong University, Qingdao 266237, China · Inspur Cloud Information Technology Co, Ltd.
The Fast Mixing Mechanism for Differential Privacy
May 28, 2026Omri Lev, Moshe Shenfeld, Vishwak Srinivasan +2Standard Differential-Privacy Stochastic Gradient Descent
Massachusetts Institute of Technology · The Hebrew University of Jerusalem
Optimal Gap-Dependent Regret for Private Stochastic Decision-Theoretic Online Learning
May 27, 2026Tommaso Cesari, Roberto ColomboniOptimal Regret Standard Differential-Privacy
School of Electrical Engineering and Computer Science University of Ottawa Ottawa, Canada · School of Mathematics University of Bristol Bristol, United Kingdom
Mind the Gap: Mixtures of Gaussians in Approximate Differential Privacy
May 27, 2026Huikang Liu, Aras Selvi, Wolfram WiesemannStandard Differential-Privacy Δ)$-Differential Privacy
1Shanghai Jiao Tong University · 2UCL School of Management · 3Imperial Business School
From Privacy to Generalization: Linear Max-Information Bounds for Differentially Private Learning Algorithms
May 25, 2026Christoph H. Lampert, Max Cairney-Leeming, Hossein ZakeriniaStandard Differential-Privacy Generalization Bounds
Institute of Science and Technology Austria (ISTA) Klosterneuburg, Austria
Deep Learning under Fractional-Order Differential Privacy
May 11, 2026Mohammad Partohaghighi, Roummel MarciaStandard Differential-Privacy Δ)$-Differential Privacy
MESA Lab, Dept. of EECS University of California, Merced Merced, CA 95343 · Dept. of Applied Mathematics University of California, Merced Merced, CA 95343
INO-SGD: Addressing Utility Imbalance under Individualized Differential Privacy
May 8, 2026Xiao Tian, Jue Fan, Rachael Hwee Ling Sim +1Standard Differential-Privacy In-Distribution
Department of Computer Science, National University of Singapore · Agency for Science, Technology and Research (A*STAR), Singapore
Trade-off Functions for DP-SGD with Subsampling based on Random Shuffling: Tight Upper and Lower Bounds
May 7, 2026Marten van Dijk, Murat Bilgehan ErtanStandard Differential-Privacy Stochastic Gradient Descent
CWI Amsterdam · ∗Affiliated with Vrije Universiteit Amsterdam.
Private Speech Classification without Collapse: Stabilized DP Training and Offline Distillation
May 4, 2026Yadi Wen, Tianxin Li, Enji Liang +2Speaker Standard Differential-Privacy
Shanxi Key Laboratory of Industrial Internet Security, Taiyuan University of Technology, China · University of Michigan, Ann Arbor, USA
Metric-Normalized Posterior Leakage (mPL): Attacker-Aligned Privacy for Joint Consumption
May 1, 2026Gaoyi Chen, Minghao Li, Weishi Shi +4Δ)$-Differential Privacy Standard Differential-Privacy
Department of Computer Science and Engineering, University of North Texas, Denton, TX, USA · Department of Electrical Engineering, University of North Texas, Denton, TX, USA
Enhanced Privacy and Communication Efficiency in Non-IID Federated Learning with Adaptive Quantization and Differential Privacy
Apr 25, 2026Emre Ardıç, Yakup GençFederated Learning Standard Differential-Privacy
Department of Computer Engineering, Gebze Technical University, Gebze, Kocaeli 41400, Türkiye
Tight Auditing of Differential Privacy in MST and AIM
Apr 20, 2026Georgi Ganev, Meenatchi Sundaram Muthu Selva Annamalai, Bogdan KulynychStandard Differential-Privacy Privacy
1UCL · 2SAS · 3Lausanne University Hospital
DPrivBench: Benchmarking LLMs' Reasoning for Differential Privacy
Apr 17, 2026Erchi Wang, Pengrun Huang, Eli Chien +4Standard Differential-Privacy Reasoning Benchmark
Halıcıo˘glu Data Science Institute, UC San Diego · Department of Computer Science and Engineering, UC San Diego · Department of Electrical Engineering, National Taiwan University +2
Efficient privacy loss accounting for subsampling and random allocation
Feb 19, 2026Vitaly Feldman, Moshe ShenfeldStandard Differential-Privacy Subsampling
Apple · The Hebrew University of Jerusalem · *Work partially done while author was an intern at Apple.
Beyond One-Size-Fits-All: Neural Networks for Differentially Private Tabular Data Synthesis
Nov 17, 2025Kai Chen, Chen Gong, Tianhao WangSynthetic Tabular Data Standard Differential-Privacy
University of Virginia Charlottesville, USA
Inclusive Federated Learning Through Compliance-Weighted Noise Allocation in Healthcare AI
May 28, 2025Santhosh Parampottupadam, Melih Coşğun, Sarthak Pati +7Federated Learning Standard Differential-Privacy
German Cancer Research Center (DKFZ), Division of Medical Image Computing, Heidelberg, Germany · Department of Computer Engineering, Bilkent University, Universiteler 06800 Çankaya/Ankara, Türkiye · Medical Research Group, MLCommons, San Francisco, CA, USA +2
Label Differential Privacy via Aggregation
Date pendingAnand Brahmbhatt, Rishi Saket, Shreyas Havaldar +3Standard Differential-Privacy Δ)$-Differential Privacy
Google Research India.

Sep 11, 2026, cs.CLJ/K move · Enter open · S save

Component-Aware Differential Privacy for Federated Multilingual Speech-LLMs

Jordi Luque, Fernando López, Aleix Sant

Scientific Research Telef´onica Innovaci´on Digital Barcelona, Spain · Telef´onica Innovaci´on Digital Universidad Aut´onoma de Madrid Madrid, Spain · Scientific Research Telef´onica Innovaci´on Digital Universitat Polit`ecnica de Catalunya Barcelona, Spain

Standard Differential-Privacy Gradient-Based Attacks Token Budget Allocation

Open paper AskPDF arXiv

Per-layer differential privacy (DP) clipping improves gradient fidelity in federated learning by allocating per-matrix clipping budgets proportional to parameter count. We show that this recipe breaks for speech large language models (speech-LLMs), when the acoustic encoder and the language decoder differ by an order of magnitude in update norm. Single-pool per-layer methods suffer \emph{cross-component budget collapse}, dragging word error rate (WER) far from flat global clipping or collapsing training entirely. When the norm imbalance is milder, adaptive single-pool methods partially recover, confirming that collapse severity scales with the inter-component norm ratio. We empirically diagnose the root cause across six per-layer methods and three speech-LLM architectures. We then propose \emph{

\alpha

-split}, a two-pool allocation that normalises encoder and LLM parameters into independent pools, and show that joint

\ell_2

sensitivity and the original

(\varepsilon,\delta)

-DP guarantee are unchanged. At architecture-calibrated

\alpha

, our method recovers WER utility compared to flat DP, while granting the encoder

4.47{\times}

tighter per-component noise protection against speaker voice-based gradient-inversion attacks at only

+2.6\%

LLM noise overhead.

Momentum

Related topics

Latest papers 25