English
Related papers

Related papers: Quantifying Privacy Risks of Masked Language Model…

200 papers

Deep learning has achieved overwhelming success, spanning from discriminative models to generative models. In particular, deep generative models have facilitated a new level of performance in a myriad of areas, ranging from media…

Machine Learning · Computer Science 2020-11-24 Dingfan Chen , Ning Yu , Yang Zhang , Mario Fritz

The potential of transformer-based LLMs risks being hindered by privacy concerns due to their reliance on extensive datasets, possibly including sensitive information. Regulatory measures like GDPR and CCPA call for using robust auditing…

Cryptography and Security · Computer Science 2024-06-25 Filippo Galli , Luca Melis , Tommaso Cucinotta

With the development of information science and technology, various industries have generated massive amounts of data, and machine learning is widely used in the analysis of big data. However, if the privacy of machine learning…

Cryptography and Security · Computer Science 2023-01-11 Jingyi Ge

Machine learning (ML) models are vulnerable to membership inference attacks (MIAs), which determine whether a given input is used for training the target model. While there have been many efforts to mitigate MIAs, they often suffer from…

Cryptography and Security · Computer Science 2023-07-06 Zitao Chen , Karthik Pattabiraman

With the rapid development of artificial intelligence, large language models (LLMs) have made remarkable advancements in natural language processing. These models are trained on vast datasets to exhibit powerful language understanding and…

Cryptography and Security · Computer Science 2025-09-22 Shang Wang , Tianqing Zhu , Bo Liu , Ming Ding , Dayong Ye , Wanlei Zhou , Philip S. Yu

Model merging is a widespread technology in large language models (LLMs) that integrates multiple task-specific LLMs into a unified one, enabling the merged model to inherit the specialized capabilities of these LLMs. Most task-specific…

Computation and Language · Computer Science 2025-02-18 Zhenyuan Guo , Yi Shi , Wenlong Meng , Chen Gong , Chengkun Wei , Wenzhi Chen

OpenLVLM-MIA is a new benchmark that highlights fundamental challenges in evaluating membership inference attacks (MIA) against large vision-language models (LVLMs). While prior work has reported high attack success rates, our analysis…

Computer Vision and Pattern Recognition · Computer Science 2025-12-03 Ryoto Miyamoto , Xin Fan , Fuyuko Kido , Tsuneo Matsumoto , Hayato Yamana

Impressive progress has been made in automated problem-solving by the collaboration of large language model (LLM) based agents. However, these automated capabilities also open avenues for malicious applications. In this paper, we study a…

Cryptography and Security · Computer Science 2026-04-16 Yuntao Du , Zitao Li , Bolin Ding , Yaliang Li , Hanshen Xiao , Jingren Zhou , Ninghui Li

Membership inference (MI) attacks exploit the fact that machine learning algorithms sometimes leak information about their training data through the learned model. In this work, we study membership inference in the white-box setting in…

Machine Learning · Computer Science 2020-06-26 Klas Leino , Matt Fredrikson

Large language models (LLMs) have demonstrated great performance across various benchmarks, showing potential as general-purpose task solvers. However, as LLMs are typically trained on vast amounts of data, a significant concern in their…

Computation and Language · Computer Science 2025-05-13 Yujuan Fu , Ozlem Uzuner , Meliha Yetisgen , Fei Xia

Memorization in Large Language Models (LLMs) poses privacy and security risks, as models may unintentionally reproduce sensitive or copyrighted data. Existing analyses focus on average-case scenarios, often neglecting the highly skewed…

Artificial Intelligence · Computer Science 2025-02-04 Hao Li , Di Huang , Ziyu Wang , Amir M. Rahmani

The prosperity of machine learning has also brought people's concerns about data privacy. Among them, inference attacks can implement privacy breaches in various MLaaS scenarios and model training/prediction phases. Specifically, inference…

Machine Learning · Computer Science 2024-06-28 Feng Wu , Lei Cui , Shaowen Yao , Shui Yu

This study explores the use of Large Language Models (LLMs) to analyze text comments from Reddit users, aiming to achieve two primary objectives: firstly, to pinpoint critical excerpts that support a predefined psychological assessment of…

Computation and Language · Computer Science 2024-02-07 Sergi Blanco-Cuaresma

Knowledge Bases (KBs) are easy to query, verifiable, and interpretable. They however scale with man-hours and high-quality data. Masked Language Models (MLMs), such as BERT, scale with computing power as well as unstructured raw text data.…

Computation and Language · Computer Science 2020-09-16 Louis Clouatre , Philippe Trempe , Amal Zouaq , Sarath Chandar

Large-scale pre-trained models are increasingly adapted to downstream tasks through a new paradigm called prompt learning. In contrast to fine-tuning, prompt learning does not update the pre-trained model's parameters. Instead, it only…

Cryptography and Security · Computer Science 2023-10-19 Yixin Wu , Rui Wen , Michael Backes , Pascal Berrang , Mathias Humbert , Yun Shen , Yang Zhang

Machine learning (ML) models have been shown to be vulnerable to Membership Inference Attacks (MIA), which infer the membership of a given data point in the target dataset by observing the prediction output of the ML model. While the key…

Machine Learning · Computer Science 2020-07-28 Shakila Mahjabin Tonni , Dinusha Vatsalan , Farhad Farokhi , Dali Kaafar , Zhigang Lu , Gioacchino Tangari

Prompt injection attacks are an emerging threat to large language models (LLMs), enabling malicious users to manipulate outputs through carefully designed inputs. Existing detection approaches often require centralizing prompt data,…

Cryptography and Security · Computer Science 2025-11-18 Hasini Jayathilaka

Membership inference attacks (MIAs) pose a significant threat to the privacy of machine learning models and are widely used as tools for privacy assessment, auditing, and machine unlearning. While prior MIA research has primarily focused on…

Machine Learning · Computer Science 2025-07-04 Zhiqi Wang , Chengyu Zhang , Yuetian Chen , Nathalie Baracaldo , Swanand Kadhe , Lei Yu

Large language models for code (LLMs4Code) rely heavily on massive training data, including sensitive data, such as cloud service credentials of the projects and personal identifiable information of the developers, raising serious privacy…

Software Engineering · Computer Science 2025-08-04 Md Nazmul Haque , Hua Yang , Zhou Yang , Bowen Xu

Membership Inference Attack (MIA) aims to determine whether a specific data sample was included in the training dataset of a target model. Traditional MIA approaches rely on shadow models to mimic target model behavior, but their…

Information Retrieval · Computer Science 2026-03-20 Li Cuihong , Huang Xiaowen , Yin Chuanhuan , Sang Jitao
‹ Prev 1 8 9 10 Next ›