Differential Privacy

Follow this topic Ask about this topic

Latest papers 193

30 papers in the last four weeks, up 100% on the four weeks before. 0.2% of all new papers.

All topics

Improved Bounds for Private and Robust Alignment
Dec 29, 2025Wenqian Weng, Yi He, Xingyu ZhouPrivacy-Preserving Language Models Corruption Robustness
Department of Electrical and Computer Engineering, Wayne State University, Detroit, USA.
Efficient Public Verification of Private ML via Regularization
Dec 3, 2025Zoë Ruha Bell, Anvith Thudi, Olive Franzese-McLaughlin +2Differential Privacy
Department of Computer Science, University of California, Berkeley, USA · Department of Computer Science, University of Toronto, Toronto, Canada · Vector Institute, Toronto, Canada
On the Gradient Complexity of Private Optimization with Private Oracles
Nov 17, 2025Michael Menart, Aleksandar NikolovQuery Complexity Oracle Complexity
Department of Computer Science, University of Toronto, Vector Institute · Department of Computer Science, University of Toronto
Beyond One-Size-Fits-All: Neural Networks for Differentially Private Tabular Data Synthesis
Nov 17, 2025Kai Chen, Chen Gong, Tianhao WangSynthetic Data Generation Synthetic Tabular Data Generation
University of Virginia Charlottesville, USA
Quantum Information Ordering and Differential Privacy
Nov 3, 2025Naqueeb Ahmad Warsi, Ayanava Dasgupta, Masahito HayashiDifferential Privacy
Indian Statistical Institute, Kolkata 700108, India. · Graduate School of Mathematics, Nagoya University, Nagoya, 464-8602, Japan.
Privacy-Preserving Generation of Clinical Narratives from Medical Terminologies
Sep 13, 2025Yuping Wu, Viktor Schlegel, Warren Del-Pinto +11Medical Report Generation Healthcare
Univeristy of Manchester · Imperial College London, Imperial Global Singapore · Khoo Teck Puat Hospital, Singapore +2
High-Dimensional Differentially Private Quantile Regression: Distributed Estimation and Statistical Inference
Aug 7, 2025Ziliang Shen, Caixing Wang, Shaoli Wang +1Quantile Regression Privacy-Preserving ML
School of Statistics and Data Science, Shanghai University of Finance and Economics · School of Statistics and Data Science, Southeast University · School of Statistics and Data Science, Shanghai University of International Business and Economics
Private Rate-Constrained Optimization with Applications to Fair Learning
May 28, 2025Mohammad Yaghini, Tudor Cebere, Michael Menart +2Primal-Dual Optimization Group Fairness
University of Toronto · Vector Institute · PreMeDICaL team +3
A Private Approximation of the 2nd-Moment Matrix of Any Subsamplable Input
May 20, 2025Bar Mahpud, Or SheffetDifferential Privacy
Faculty of Engineering Bar-Ilan University Israel
A Differentially Private Weighted Empirical Risk Minimization Procedure and its Application to Outcome Weighted Learning
Jul 24, 2023Spencer Giddens, Yiwang Zhou, Kevin R. Krull +3Privacy-Preserving ML Differential Privacy
University of Notre Dame, Notre Dame, IN 46556, USA · St. Jude Children’s Research Hospital, Memphis, TN 38138, USA · University of Michigan, Ann Arbor, MI 48109, USA
Label Differential Privacy via Aggregation
Date pendingAnand Brahmbhatt, Rishi Saket, Shreyas Havaldar +3Privacy-Preserving ML Differential Privacy
Google Research India.
Protect Your Score: Contact Tracing With Differential Privacy Guarantees
Date pendingRob Romijnders, Christos Louizos, Yuki M. Asano +1Differential Privacy Privacy-Utility Tradeoff
University of Amsterdam · Qualcomm AI research
DNA: Differentially private Neural Augmentation for contact tracing
Date pendingRob Romijnders, Christos Louizos, Yuki M. Asano +1Privacy-Preserving ML Differential Privacy
University of Amsterdam · Qualcomm AI research

Dec 29, 2025, cs.LGJ/K move · Enter open · S save

Improved Bounds for Private and Robust Alignment

Wenqian Weng, Yi He, Xingyu Zhou

Department of Electrical and Computer Engineering, Wayne State University, Detroit, USA.

Privacy-Preserving Language Models Corruption Robustness Differential Privacy Preference Alignment

Open paper AskPDF arXiv

In this paper, we study the private and robust alignment of language models from a theoretical perspective by establishing upper bounds on the suboptimality gap in both offline and online settings. We consider preference labels subject to privacy constraints and/or adversarial corruption, and analyze two distinct interplays between them: privacy-first and corruption-first. For the privacy-only setting, we show that log loss with an MLE-style algorithm achieves near-optimal rates, in contrast to conventional wisdom. For the joint privacy-and-corruption setting, we first demonstrate that existing offline algorithms in fact provide stronger guarantees -- simultaneously in terms of corruption level and privacy parameters -- than previously known, which further yields improved bounds in the corruption-only regime. In addition, we also present the first set of results for private and robust online alignment. Our results are enabled by new uniform convergence guarantees for log loss and square loss under privacy and corruption, which we believe have broad applicability across learning theory and statistics.