English
Related papers

Related papers: Membership Inference on LLMs in the Wild

200 papers

In membership inference attacks (MIAs), an adversary observes the predictions of a model to determine whether a sample is part of the model's training data. Existing MIA defenses conceal the presence of a target sample through strong…

Cryptography and Security · Computer Science 2022-03-03 Ismat Jarin , Birhanu Eshete

The integration of large language models with external content has enabled applications such as Microsoft Copilot but also introduced vulnerabilities to indirect prompt injection attacks. In these attacks, malicious instructions embedded…

Computation and Language · Computer Science 2025-01-28 Jingwei Yi , Yueqi Xie , Bin Zhu , Emre Kiciman , Guangzhong Sun , Xing Xie , Fangzhao Wu

This paper presents how to leak private information from a wireless signal classifier by launching an over-the-air membership inference attack (MIA). As machine learning (ML) algorithms are used to process wireless signals to make decisions…

Signal Processing · Electrical Eng. & Systems 2020-06-26 Yi Shi , Kemal Davaslioglu , Yalin E. Sagduyu

Large language models (LLMs) have demonstrated significant utility in a wide range of applications; however, their deployment is plagued by security vulnerabilities, notably jailbreak attacks. These attacks manipulate LLMs to generate…

Computation and Language · Computer Science 2025-03-12 Wenlong Meng , Fan Zhang , Wendao Yao , Zhenyuan Guo , Yuwei Li , Chengkun Wei , Wenzhi Chen

This report summarizes all the MIA experiments (Membership Inference Attacks) of the Embedding Attack Project, including threat models, experimental setup, experimental results, findings and discussion. Current results cover the evaluation…

Machine Learning · Computer Science 2024-01-26 Jiameng Pu , Zafar Takhirov

Membership inference attacks are used as a key tool for disclosure auditing. They aim to infer whether an individual record was used to train a model. While such evaluations are useful to demonstrate risk, they are computationally expensive…

Machine Learning · Computer Science 2024-12-20 Anshuman Suri , Xiao Zhang , David Evans

Most membership inference attacks (MIAs) against Large Language Models (LLMs) rely on global signals, like average loss, to identify training data. This approach, however, dilutes the subtle, localized signals of memorization, reducing…

Computation and Language · Computer Science 2026-03-09 Yuetian Chen , Yuntao Du , Kaiyuan Zhang , Ashish Kundu , Charles Fleming , Bruno Ribeiro , Ninghui Li

The increased availability of genetic data has transformed genomics research, but raised many privacy concerns regarding its handling due to its sensitive nature. This work explores the use of language models (LMs) for the generation of…

Cryptography and Security · Computer Science 2025-12-19 Asia Belfiore , Jonathan Passerat-Palmbach , Dmitrii Usynin

Transfer learning has been widely studied and gained increasing popularity to improve the accuracy of machine learning models by transferring some knowledge acquired in different training. However, no prior work has pointed out that…

Cryptography and Security · Computer Science 2023-07-19 Seira Hidano , Takao Murakami , Yusuke Kawamoto

Model inversion attacks (MIAs) aim to recover private data from inaccessible training sets of deep learning models, posing a privacy threat. MIAs primarily focus on the white-box scenario where attackers have full access to the model's…

Artificial Intelligence · Computer Science 2024-03-07 Rongke Liu , Dong Wang , Yizhi Ren , Zhen Wang , Kaitian Guo , Qianqian Qin , Xiaolei Liu

The pretraining and fine-tuning approach has become the leading technique for various NLP applications. However, recent studies reveal that fine-tuning data, due to their sensitive nature, domain-specific characteristics, and…

Computation and Language · Computer Science 2024-11-13 Qian Sun , Hanpeng Wu , Xi Sheryl Zhang

The high cost of model training makes it increasingly desirable to develop techniques for unlearning. These techniques seek to remove the influence of a training example without having to retrain the model from scratch. Intuitively, once a…

Machine Learning · Computer Science 2024-05-22 Jamie Hayes , Ilia Shumailov , Eleni Triantafillou , Amr Khalifa , Nicolas Papernot

Large Language Models (LLMs) embed sensitive, human-generated data, prompting the need for unlearning methods. Although certified unlearning offers strong privacy guarantees, its restrictive assumptions make it unsuitable for LLMs, giving…

Machine Learning · Computer Science 2025-06-03 Rongzhe Wei , Mufei Li , Mohsen Ghassemi , Eleonora Kreačić , Yifan Li , Xiang Yue , Bo Li , Vamsi K. Potluru , Pan Li , Eli Chien

Recommender systems have been successfully applied in many applications. Nonetheless, recent studies demonstrate that recommender systems are vulnerable to membership inference attacks (MIAs), leading to the leakage of users' membership…

Cryptography and Security · Computer Science 2024-05-14 Xiaoxiao Chi , Xuyun Zhang , Yan Wang , Lianyong Qi , Amin Beheshti , Xiaolong Xu , Kim-Kwang Raymond Choo , Shuo Wang , Hongsheng Hu

Membership inference attacks (MIAs) are becoming standard tools for auditing the privacy of machine learning models. The leading attacks -- LiRA (Carlini et al., 2022) and RMIA (Zarifzadeh et al., 2024) -- appear to use distinct scoring…

Machine Learning · Computer Science 2026-03-13 Rickard Brännvall

Federated Large Language Models (FedLLMs) enable multiple parties to collaboratively fine-tune LLMs without sharing raw data, addressing challenges of limited resources and privacy concerns. Despite data localization, shared gradients can…

Machine Learning · Computer Science 2026-04-24 Guilin Deng , Silong Chen , Yuchuan Luo , Yi Liu , Songlei Wang , Zhiping Cai , Lin Liu , Xiaohua Jia , Shaojing Fu

A number of recent works have demonstrated that API access to machine learning models leaks information about the dataset records used to train the models. Further, the work of \cite{somesh-overfit} shows that such membership inference…

Cryptography and Security · Computer Science 2019-10-15 Benjamin Zi Hao Zhao , Hassan Jameel Asghar , Raghav Bhaskar , Mohamed Ali Kaafar

A fundamental issue in deep learning has been adversarial robustness. As these systems have scaled, such issues have persisted. Currently, large language models (LLMs) with billions of parameters suffer from adversarial attacks just like…

Machine Learning · Computer Science 2025-02-11 Brian Formento , Chuan Sheng Foo , See-Kiong Ng

Membership inference attack (MIA) poses a significant privacy threat in federated learning (FL) as it allows adversaries to determine whether a client's private dataset contains a specific data sample. While defenses against membership…

Machine Learning · Computer Science 2026-02-10 Quan Minh Nguyen , Min-Seon Kim , Hoang M. Ngo , Trong Nghia Hoang , Hyuk-Yoon Kwon , My T. Thai

Masked Diffusion Language Models MDLMs replace autoregressive generation with iterative demasking and their privacy properties are largely unstudied. We study membership inference attacks MIA on fine tuned MDLMs and show they are…

Machine Learning · Computer Science 2026-05-20 Shailesh Kasivelrajan