Cover Image

Mind the Privacy Unit! User-Level Differential Privacy for Language Model Fine-Tuning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Chua, Lynn, Ghazi, Badih, Huang, Yangsibo, Kamath, Pritish, Kumar, Ravi, Liu, Daogao, Manurangsi, Pasin, Sinha, Amer, Zhang, Chiyuan
Format:	Preprint
Published:	2024
Subjects:	Computation and Language Cryptography and Security Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Unlearn and Burn: Adversarial Machine Unlearning Requests Destroy Model Accuracy
by: Huang, Yangsibo, et al.
Published: (2024)

Scalable DP-SGD: Shuffling vs. Poisson Subsampling
by: Chua, Lynn, et al.
Published: (2024)

How Private are DP-SGD Implementations?
by: Chua, Lynn, et al.
Published: (2024)

On Computing Pairwise Statistics with Local Differential Privacy
by: Ghazi, Badih, et al.
Published: (2024)

Individualized Privacy Accounting via Subsampling with Applications in Combinatorial Optimization
by: Ghazi, Badih, et al.
Published: (2024)

Private Hyperparameter Tuning with Ex-Post Guarantee
by: Ghazi, Badih, et al.
Published: (2025)

Linear-Time User-Level DP-SCO via Robust Statistics
by: Ghazi, Badih, et al.
Published: (2025)

Balls-and-Bins Sampling for DP-SGD
by: Chua, Lynn, et al.
Published: (2024)

On Convex Optimization with Semi-Sensitive Features
by: Ghazi, Badih, et al.
Published: (2024)

Denoising the US Census: Succinct Block Hierarchical Regression
by: Ghazi, Badih, et al.
Published: (2026)

Differential Privacy on Trust Graphs
by: Ghazi, Badih, et al.
Published: (2024)

On the Differential Privacy and Interactivity of Privacy Sandbox Reports
by: Ghazi, Badih, et al.
Published: (2024)

Training Differentially Private Ad Prediction Models with Semi-Sensitive Features
by: Chua, Lynn, et al.
Published: (2024)

Computational Hardness of Private Coreset
by: Ghazi, Badih, et al.
Published: (2026)

Urania: Differentially Private Insights into AI Use
by: Liu, Daogao, et al.
Published: (2025)

VaultGemma: A Differentially Private Gemma Model
by: Sinha, Amer, et al.
Published: (2025)

Crosslingual Capabilities and Knowledge Barriers in Multilingual Large Language Models
by: Chua, Lynn, et al.
Published: (2024)

Convex Optimization with Local Label Differential Privacy: Tight Bounds in All Privacy Regimes
by: Chua, Lynn, et al.
Published: (2026)

Infinitely Divisible Noise for Differential Privacy: Nearly Optimal Error in the High $\varepsilon$ Regime
by: Harrison, Charlie, et al.
Published: (2025)

Scaling Laws for Differentially Private Language Models
by: McKenna, Ryan, et al.
Published: (2025)

Differentially Private Ad Conversion Measurement
by: Delaney, John, et al.
Published: (2024)

Differentially Private Optimization with Sparse Gradients
by: Ghazi, Badih, et al.
Published: (2024)

Empirical Privacy Variance
by: Hu, Yuzheng, et al.
Published: (2025)

Improved Lower Bound for Differentially Private Facility Location
by: Manurangsi, Pasin
Published: (2024)

Exact zCDP Characterizations for Fundamental Differentially Private Mechanisms
by: Harrison, Charlie, et al.
Published: (2025)

Differentially Private Fair Division
by: Manurangsi, Pasin, et al.
Published: (2022)

Optimal partition selection with Rényi differential privacy
by: Harrison, Charlie, et al.
Published: (2026)

The Price of Privacy For Approximating Max-CSP
by: Dharangutte, Prathamesh, et al.
Published: (2026)

Scaling Embedding Layers in Language Models
by: Yu, Da, et al.
Published: (2025)

Fine-Tuning Large Language Models with User-Level Differential Privacy
by: Charles, Zachary, et al.
Published: (2024)

Protecting User Prompts Via Character-Level Differential Privacy
by: Arachchige, Shashie Dilhara Batan, et al.
Published: (2026)

Metric Differential Privacy at the User-Level Via the Earth Mover's Distance
by: Imola, Jacob, et al.
Published: (2024)

Privacy Filters are Captured by Residues: A Characterization of Free Natural Filters and the Cost of Adaptivity
by: Regehr, Matthew, et al.
Published: (2026)

Nearly-Optimal Private Selection via Gaussian Mechanism
by: Leeman, Ethan, et al.
Published: (2025)

Improved Differentially Private Algorithms for Rank Aggregation
by: Hillebrand, Quentin, et al.
Published: (2025)

The Discrete Gaussian for Differential Privacy
by: Canonne, Clément L., et al.
Published: (2020)

When FinTech Meets Privacy: Securing Financial LLMs with Differential Private Fine-Tuning
by: Zhu, Sichen, et al.
Published: (2025)

PrivShape: Extracting Shapes in Time Series under User-Level Local Differential Privacy
by: Mao, Yulian, et al.
Published: (2024)

Revisiting Hyperparameter Tuning with Differential Privacy
by: Ding, Youlong, et al.
Published: (2022)

Practitioners' Perspectives on a Differential Privacy Deployment Registry
by: Nanayakkara, Priyanka, et al.
Published: (2025)