English
Related papers

Related papers: Differentially Private Zeroth-Order Methods for Sc…

200 papers

Fine-tuning large language models (LLMs) with classic first-order optimizers entails prohibitive GPU memory due to the backpropagation process. Recent works have turned to zeroth-order optimizers for fine-tuning, which save substantial…

Machine Learning · Computer Science 2025-02-19 Yanjun Zhao , Sizhe Dang , Haishan Ye , Guang Dai , Yi Qian , Ivor W. Tsang

Fine-tuning vision language models (VLMs) has achieved remarkable performance across various downstream tasks; yet, it requires access to model gradients through backpropagation (BP), making them unsuitable for memory-constrained,…

Machine Learning · Computer Science 2025-10-27 Yifan Yang , Zhen Zhang , Rupak Vignesh Swaminathan , Jing Liu , Nathan Susanj , Zheng Zhang

We study the problem of differentially private (DP) fine-tuning of large pre-trained models -- a recent privacy-preserving approach suitable for solving downstream tasks with sensitive data. Existing work has demonstrated that high accuracy…

Machine Learning · Computer Science 2024-06-21 Zhiqi Bu , Yu-Xiang Wang , Sheng Zha , George Karypis

Open-source Large Language Models (LLMs) often employ safety alignment methods to resist harmful instructions. However, recent research shows that maliciously fine-tuning these LLMs on harmful data can easily bypass these safeguards. To…

Cryptography and Security · Computer Science 2025-07-30 Zixuan Chen , Weikai Lu , Xin Lin , Ziqian Zeng

Graphs offer unique insights into relationships between entities, complementing data modalities like text and images and enabling AI models to extend their capabilities beyond traditional tasks. However, learning from graphs often involves…

Machine Learning · Computer Science 2025-09-18 Haoteng Yin , Rongzhe Wei , Eli Chien , Pan Li

Large language models (LLMs) frequently memorize sensitive or personal information, raising significant privacy concerns. Existing variants of differential privacy stochastic gradient descent (DPSGD) inject uniform noise into every gradient…

Machine Learning · Computer Science 2025-09-30 Manjiang Yu , Priyanka Singh , Xue Li , Yang Cao

In the domain of deep learning, the challenge of protecting sensitive data while maintaining model utility is significant. Traditional Differential Privacy (DP) techniques such as Differentially Private Stochastic Gradient Descent (DP-SGD)…

Machine Learning · Computer Science 2024-11-06 Tao Huang , Qingyu Huang , Xin Shi , Jiayang Meng , Guolong Zheng , Xu Yang , Xun Yi

In recent years, large language models (LLMs) have significantly advanced the field of natural language processing (NLP). By fine-tuning LLMs with data from specific scenarios, these foundation models can better adapt to various downstream…

Computation and Language · Computer Science 2024-11-05 Jiaqi Wu , Simin Chen , Yuzhe Yang , Yijiang Li , Shiyue Hou , Rui Jing , Zehua Wang , Wei Chen , Zijian Tian

Protecting large language models from privacy leakage is becoming increasingly crucial with their wide adoption in real-world products. Yet applying differential privacy (DP), a canonical notion with provable privacy guarantees for machine…

Computation and Language · Computer Science 2022-10-28 Weiyan Shi , Ryan Shea , Si Chen , Chiyuan Zhang , Ruoxi Jia , Zhou Yu

Supervised Fine-Tuning (SFT) Large Language Models (LLM) fundamentally rely on high-quality training data. While data selection and data synthesis are two common strategies to improve data quality, existing approaches often face limitations…

Computation and Language · Computer Science 2025-10-23 Zinan Tang , Xin Gao , Qizhi Pei , Zhuoshi Pan , Mengzhang Cai , Jiang Wu , Conghui He , Lijun Wu

Large language models often require fine-tuning to better align their behavior with user intent at deployment. Existing approaches are commonly divided into online and offline paradigms. Online methods, such as RL-based alignment, can…

Machine Learning · Computer Science 2026-05-19 Shijun Li , Kaiwen Dong , Xiang Gao , Joydeep Ghosh

This study evaluates Direct Preference Optimization (DPO) and its variants for aligning Large Language Models (LLMs) with human preferences, testing three configurations: (1) with Supervised Fine Tuning (SFT), (2) without SFT, and (3)…

Computation and Language · Computer Science 2025-02-11 Amir Saeidi , Shivanshu Verma , Md Nayem Uddin , Chitta Baral

Lowering the memory requirement in full-parameter training on large models has become a hot research area. MeZO fine-tunes the large language models (LLMs) by just forward passes in a zeroth-order SGD optimizer (ZO-SGD), demonstrating…

Machine Learning · Computer Science 2023-12-27 Shuoran Jiang , Qingcai Chen , Youchen Pan , Yang Xiang , Yukang Lin , Xiangping Wu , Chuanyi Liu , Xiaobao Song

This study addresses the issues of privacy protection and efficiency in instruction fine-tuning of large-scale language models by proposing a parameter-efficient method that integrates differential privacy noise allocation with gradient…

Computation and Language · Computer Science 2025-12-09 Yulin Huang , Yaxuan Luan , Jinxu Guo , Xiangchen Song , Yuchen Liu

We study the inherent trade-offs in minimizing privacy risks and maximizing utility, while maintaining high computational efficiency, when fine-tuning large language models (LLMs). A number of recent works in privacy research have attempted…

Artificial Intelligence · Computer Science 2026-02-10 Soumi Das , Camila Kolling , Mohammad Aflah Khan , Mahsa Amani , Bishwamittra Ghosh , Qinyuan Wu , Till Speicher , Krishna P. Gummadi

Zeroth-order (ZO) optimization has become increasingly popular and important in fine-tuning large language models (LLMs), especially on edge devices due to its ability to adjust the model to local data without the need for memory-intensive…

Machine Learning · Computer Science 2026-05-18 Jiahe Chen , Ziye Ma

Tuning the hyperparameters of differentially private (DP) machine learning (ML) algorithms often requires use of sensitive data and this may leak private information via hyperparameter values. Recently, Papernot and Steinke (2022) proposed…

Machine Learning · Computer Science 2024-02-14 Antti Koskela , Tejas Kulkarni

Classical optimization theory establishes that zeroth-order (ZO) algorithms suffer from a dimension-dependent slowdown, with convergence rates typically scaling with the model dimension compared to first-order methods. However, in contrast…

Machine Learning · Computer Science 2026-05-06 Zhe Li , Bicheng Ying , Zidong Liu , Haibo Yang

Differential privacy (DP) is a privacy-preserving paradigm that protects the training data when training deep learning models. Critically, the performance of models is determined by the training hyperparameters, especially those of the…

Machine Learning · Computer Science 2025-03-04 Zhiqi Bu , Ruixuan Liu

Differentially private SGD (DPSGD) enables privacy-preserving training of language models, but often reduces utility, diversity, and linguistic quality. We introduce DPRefine, a three-phase method that initializes a model using data…

Machine Learning · Computer Science 2024-10-24 Ivoline C. Ngong , Joseph P. Near , Niloofar Mireshghallah