paper-with-me

홈 › Papers

DPrivBench: Benchmarking LLMs' Reasoning for Differential Privacy

2026-04-17 · Erchi Wang, Pengrun Huang, Eli Chien, Om Thakkar, Kamalika Chaudhuri, Yu-Xiang Wang, Ruihan Wu arxiv

Differential privacy (DP) has a wide range of applications for protecting data privacy, but designing and verifying DP algorithms requires expert-level reasoning, creating a high barrier for non-expert practitioners. Prior works either rely on specialized verification languages that demand substantial domain expertise or remain semi-automated and require human-in-the-loop guidance. In this work, we investigate whether large language models (LLMs) can automate DP reasoning. We introduce DPrivBench, a benchmark in which each instance asks whether a function or algorithm satisfies a stated DP guarantee under specified assumptions. The benchmark is carefully designed to cover a broad range of DP topics, span diverse difficulty levels, and resist shortcut reasoning through trivial pattern matching. Experiments show that while the strongest models handle textbook mechanisms well, all models struggle with advanced algorithms, revealing substantial gaps in current DP reasoning capabilities. Through further analytic study and failure-mode analysis, we identify several promising directions for improving automated DP reasoning. Our benchmark provides a solid foundation for developing and evaluating such methods, and complements existing benchmarks for mathematical reasoning.

📄 PDF Abstract BibTeX arXiv:2604.15851

Code (0)

등록된 구현이 없습니다.

Tasks

Mathematical Reasoning

Similar Papers 제목 키워드 기반

Benchmarking Differentially Private Residual Networks for Medical Imagery

2020-05-27 · Sahib Singh, Harshvardhan Sikka, Sasikanth Kotti, Andrew Trask

In this paper we measure the effectiveness of $\epsilon$-Differential Privacy (DP) when applied to medical imaging. We compare two robust differential privacy mechanisms: Local-DP and DP-SGD and benchmark their performan…

Benchmarking

Benchmarking Empirical Privacy Protection for Adaptations of Large Language Models

2026-06-08 · Bartłomiej Marek, Lorenzo Rossi, Vincent Hanke, Xun Wang 외 arxiv

Recent work has applied differential privacy (DP) to adapt large language models (LLMs) for sensitive applications, offering theoretical guarantees. However, its practical effectiveness remains unclear, partly due to LLM…

parameter-efficient fine-tuning

A Comparative Analysis of Word-Level Metric Differential Privacy: Benchmarking The Privacy-Utility Trade-off

2024-04-04 · Stephen Meisenbacher, Nihildev Nandakumar, Alexandra Klymenko, Florian Matthes

The application of Differential Privacy to Natural Language Processing techniques has emerged in relevance in recent years, with an increasing number of studies published in established NLP outlets. In particular, the ad…

Benchmarking

Evaluating Differentially Private Generation of Domain-Specific Text

2025-08-28 · Yidan Sun, Viktor Schlegel, Srinivasan Nandakumar, Iqra Zahid 외 arxiv

Generative AI offers transformative potential for high-stakes domains such as healthcare and finance, yet privacy and regulatory barriers hinder the use of real-world data. To address this, differentially private synthet…

Synthetic Data Generation

A Statistical Viewpoint on Differential Privacy: Hypothesis Testing, Representation and Blackwell's Theorem

2024-09-14 · Weijie J. Su

Differential privacy is widely considered the formal privacy for privacy-preserving data analysis due to its robust and rigorous guarantees, with increasingly broad adoption in public services, academia, and industry. De…

InformativenessPrivacy Preserving