English
Related papers

Related papers: Deduplicating Training Data Mitigates Privacy Risk…

200 papers

Recent research shows that large language models are susceptible to privacy attacks that infer aspects of the training data. However, it is unclear if simpler generative models, like topic models, share similar vulnerabilities. In this…

Cryptography and Security · Computer Science 2024-09-24 Nico Manzonelli , Wanrong Zhang , Salil Vadhan

Frontier AI systems are making transformative impacts across society, but such benefits are not without costs: models trained on web-scale datasets containing personal and private data raise profound concerns about data privacy and…

Computer Vision and Pattern Recognition · Computer Science 2024-07-26 Sunny Duan , Mikail Khona , Abhiram Iyer , Rylan Schaeffer , Ila R Fiete

Deduplication is a vital preprocessing step that enhances machine learning model performance and saves training time and energy. However, enhancing federated learning through deduplication poses challenges, especially regarding scalability…

Cryptography and Security · Computer Science 2024-12-05 Aydin Abadi , Vishnu Asutosh Dasu , Sumanta Sarkar

Studying data memorization in neural language models helps us understand the risks (e.g., to privacy or copyright) associated with models regurgitating training data and aids in the development of countermeasures. Many prior works -- and…

Privacy attacks on machine learning models aim to identify the data that is used to train such models. Such attacks, traditionally, are studied on static models that are trained once and are accessible by the adversary. Motivated to meet…

Machine Learning · Computer Science 2022-02-09 Ji Gao , Sanjam Garg , Mohammad Mahmoody , Prashant Nalini Vasudevan

Large language models pretrained on a huge amount of data capture rich knowledge and information in the training data. The ability of data memorization and regurgitation in pretrained language models, revealed in previous studies, brings…

Cryptography and Security · Computer Science 2023-12-06 Xinwei Wu , Junzhuo Li , Minghui Xu , Weilong Dong , Shuangzhi Wu , Chao Bian , Deyi Xiong

Recent research has shown that structured machine learning models such as tree ensembles are vulnerable to privacy attacks targeting their training data. To mitigate these risks, differential privacy (DP) has become a widely adopted…

Machine Learning · Computer Science 2026-01-07 Alice Gorgé , Julien Ferry , Sébastien Gambs , Thibaut Vidal

The last decade has seen a rise of Deep Learning with its applications ranging across diverse domains. But usually, the datasets used to drive these systems contain data which is highly confidential and sensitive. Though, Deep Learning…

Cryptography and Security · Computer Science 2022-12-09 Vishal Jignesh Gandhi , Sanchit Shokeen , Saloni Koshti

We present new mechanisms for \emph{label differential privacy}, a relaxation of differentially private machine learning that only protects the privacy of the labels in the training set. Our mechanisms cluster the examples in the training…

Machine Learning · Computer Science 2021-10-06 Hossein Esfandiari , Vahab Mirrokni , Umar Syed , Sergei Vassilvitskii

Reconstruction attacks allow an adversary to regenerate data samples of the training set using access to only a trained model. It has been recently shown that simple heuristics can reconstruct data samples from language models, making this…

Machine Learning · Computer Science 2022-02-16 Pierre Stock , Igor Shilov , Ilya Mironov , Alexandre Sablayrolles

Deep Learning has recently become hugely popular in machine learning, providing significant improvements in classification accuracy in the presence of highly-structured and large databases. Researchers have also considered privacy…

Cryptography and Security · Computer Science 2017-09-15 Briland Hitaj , Giuseppe Ateniese , Fernando Perez-Cruz

As the deployment of pre-trained language models (PLMs) expands, pressing security concerns have arisen regarding the potential for malicious extraction of training data, posing a threat to data privacy. This study is the first to provide a…

Computation and Language · Computer Science 2023-05-26 Shotaro Ishihara

Recent developments in language modeling have increased their use in various applications and domains. Language models, often trained on sensitive data, can memorize and disclose this information during privacy attacks, raising concerns…

Computation and Language · Computer Science 2025-08-22 Pritilata Saha , Abhirup Sinha

One of the big challenges in machine learning applications is that training data can be different from the real-world data faced by the algorithm. In language modeling, users' language (e.g. in private messaging) could change in a year and…

Computation and Language · Computer Science 2018-03-07 Vadim Popov , Mikhail Kudinov , Irina Piontkovskaya , Petr Vytovtov , Alex Nevidomsky

Although language models (LMs) demonstrate exceptional capabilities on various tasks, they are potentially vulnerable to extraction attacks, which represent a significant privacy risk. To mitigate the privacy concerns of LMs, machine…

Computation and Language · Computer Science 2024-06-21 Dohyun Lee , Daniel Rim , Minseok Choi , Jaegul Choo

The effectiveness of large language models (LLMs) is often hindered by duplicated data in their extensive pre-training datasets. Current approaches primarily focus on detecting and removing duplicates, which risks the loss of valuable…

Computation and Language · Computer Science 2024-07-10 Nan He , Weichen Xiong , Hanwen Liu , Yi Liao , Lei Ding , Kai Zhang , Guohua Tang , Xiao Han , Wei Yang

Training data memorization in language models impacts model capability (generalization) and safety (privacy risk). This paper focuses on analyzing prompts' impact on detecting the memorization of 6 masked language model-based named entity…

Computation and Language · Computer Science 2024-05-07 Yuxi Xia , Anastasiia Sedova , Pedro Henrique Luz de Araujo , Vasiliki Kougia , Lisa Nußbaumer , Benjamin Roth

Recent research has shown that language models have a tendency to memorize rare or unique sequences in the training corpora which can thus leak sensitive attributes of user data. We employ a teacher-student framework and propose a novel…

Computation and Language · Computer Science 2022-10-14 Zhe Liu , Xuedong Zhang , Fuchun Peng

Large language models, trained on massive corpora, are prone to verbatim memorization of training data, creating significant privacy and copyright risks. While previous works have proposed various definitions for memorization, many exhibit…

Computation and Language · Computer Science 2025-11-27 Trung Cuong Dang , David Mohaisen

This paper studies model-inversion attacks, in which the access to a model is abused to infer information about the training data. Since its first introduction, such attacks have raised serious concerns given that training data usually…

Machine Learning · Computer Science 2020-04-21 Yuheng Zhang , Ruoxi Jia , Hengzhi Pei , Wenxiao Wang , Bo Li , Dawn Song