English
Related papers

Related papers: Does Data Contamination Detection Work (Well) for …

200 papers

Recent statements about the impressive capabilities of large language models (LLMs) are usually supported by evaluating on open-access benchmarks. Considering the vast size and wide-ranging sources of LLMs' training data, it could…

Computation and Language · Computer Science 2024-06-03 Yihong Dong , Xue Jiang , Huanyu Liu , Zhi Jin , Bin Gu , Mengfei Yang , Ge Li

Large Multimodal Language Models (MLLMs) are emerging as one of the foundational tools in an expanding range of applications. Consequently, understanding training-data leakage in these systems is increasingly critical. Log-probability-based…

Cryptography and Security · Computer Science 2026-05-22 Ziyi Tong , Feifei Sun , Le Minh Nguyen

We present the first systematic Membership Inference Attack (MIA) evaluation of Large Audio Language Models (LALMs). As audio encodes non-semantic information, it induces severe train and test distribution shifts and can lead to spurious…

Sound · Computer Science 2026-03-31 Jia-Kai Dong , Yu-Xiang Lin , Hung-Yi Lee

The lack of data transparency in Large Language Models (LLMs) has highlighted the importance of Membership Inference Attack (MIA), which differentiates trained (member) and untrained (non-member) data. Though it shows success in previous…

Computation and Language · Computer Science 2024-12-19 Bowen Chen , Namgi Han , Yusuke Miyao

Whether LLMs memorize their training data and what this means, from measuring privacy leakage to detecting copyright violations, has become a rapidly growing area of research. In the last few months, more than 10 new methods have been…

Computation and Language · Computer Science 2025-03-10 Matthieu Meeus , Igor Shilov , Shubham Jain , Manuel Faysse , Marek Rei , Yves-Alexandre de Montjoye

Public benchmarks play an essential role in the evaluation of large language models. However, data contamination can lead to inflated performance, rendering them unreliable for model comparison. It is therefore crucial to detect…

Computation and Language · Computer Science 2024-05-28 Jasper Dekoninck , Mark Niklas Müller , Martin Vechev

Membership Inference Attacks (MIAs) infer whether a data point is in the training data of a machine learning model. It is a threat while being in the training data is private information of a data point. MIA correctly infers some data…

Cryptography and Security · Computer Science 2022-10-31 Mauro Conti , Jiaxin Li , Stjepan Picek

The lifecycle of large language models (LLMs) is far more complex than that of traditional machine learning models, involving multiple training stages, diverse data sources, and varied inference methods. While prior research on data…

Cryptography and Security · Computer Science 2025-02-21 Pengfei He , Yue Xing , Han Xu , Zhen Xiang , Jiliang Tang

Recent studies apply psychometric questionnaires to Large Language Models (LLMs) to assess high-level psychological constructs such as values, personality, moral foundations, and dark traits. Although prior work has raised concerns about…

Computation and Language · Computer Science 2026-02-02 Jongwook Han , Woojung Song , Jonggeun Lee , Yohan Jo

Hampering the interpretation of benchmark scores, evaluation data contamination has become a growing concern in the evaluation of LLMs, and an active area of research studies its effects. While evaluation data contamination is easily…

Computation and Language · Computer Science 2024-11-07 Aaditya K. Singh , Muhammed Yusuf Kocyigit , Andrew Poulton , David Esiobu , Maria Lomeli , Gergely Szilvasy , Dieuwke Hupkes

Data contamination in model evaluation has become increasingly prevalent with the growing popularity of large language models. It allows models to "cheat" via memorisation instead of displaying true capabilities. Therefore, contamination…

Computation and Language · Computer Science 2024-01-30 Yucheng Li , Frank Guerin , Chenghua Lin

Data contamination -- the accidental consumption of evaluation examples within the pre-training data -- can undermine the validity of evaluation benchmarks. In this paper, we present a rigorous analysis of the effects of contamination on…

Computation and Language · Computer Science 2025-02-03 Muhammed Yusuf Kocyigit , Eleftheria Briakou , Daniel Deutsch , Jiaming Luo , Colin Cherry , Markus Freitag

Membership Inference Attacks (MIAs) aim to predict whether a data sample belongs to the model's training set or not. Although prior research has extensively explored MIAs in Large Language Models (LLMs), they typically require accessing to…

Cryptography and Security · Computer Science 2025-02-27 Yu He , Boheng Li , Liu Liu , Zhongjie Ba , Wei Dong , Yiming Li , Zhan Qin , Kui Ren , Chun Chen

The increasing parameters and expansive dataset of large language models (LLMs) highlight the urgent demand for a technical solution to audit the underlying privacy risks and copyright issues associated with LLMs. Existing studies have…

Computation and Language · Computer Science 2024-12-30 Wenjie Fu , Huandong Wang , Chen Gao , Guanghua Liu , Yong Li , Tao Jiang

Membership Inference Attacks (MIAs) have recently been employed to determine whether a specific text was part of the pre-training data of Large Language Models (LLMs). However, existing methods often misinfer non-members as members, leading…

Machine Learning · Computer Science 2025-05-07 Saleh Zare Zade , Yao Qiang , Xiangyu Zhou , Hui Zhu , Mohammad Amin Roshani , Prashant Khanduri , Dongxiao Zhu

Large Language Models (LLMs) have become the predominant paradigm in NLP, advancing both research and industry. As model sizes and pretraining data grow, concerns about Pretraining Data Exposure (PDE) increase due to the scale and opacity…

Computation and Language · Computer Science 2026-05-27 Ziyi Tong , Feifei Sun , Le Minh Nguyen

Large vision-language models (LVLMs) have demonstrated outstanding performance in many downstream tasks. However, LVLMs are trained on large-scale datasets, which can pose privacy risks if training images contain sensitive information.…

Computer Vision and Pattern Recognition · Computer Science 2026-02-05 Zongyu Wu , Minhua Lin , Zhiwei Zhang , Fali Wang , Xianren Zhang , Xiang Zhang , Suhang Wang

Large language models (LLMs) can leak sensitive training data through memorization and membership inference attacks. Prior work has primarily focused on strong adversarial assumptions, including attacker access to entire samples or long,…

Machine Learning · Computer Science 2025-05-21 Lucas Rosenblatt , Bin Han , Robert Wolfe , Bill Howe

The memorization of training data in large language models (LLMs) poses significant privacy and copyright concerns. Existing data extraction methods, particularly heuristic-based divergence attacks, often exhibit limited success and offer…

Computation and Language · Computer Science 2025-11-11 Myeongseob Ko , Nikhil Reddy Billa , Adam Nguyen , Charles Fleming , Ming Jin , Ruoxi Jia

The advancement of large language models (LLMs) relies on evaluation using public benchmarks, but data contamination can lead to overestimated performance. Previous researches focus on detecting contamination by determining whether the…

Computation and Language · Computer Science 2024-09-24 Shangqing Tu , Kejian Zhu , Yushi Bai , Zijun Yao , Lei Hou , Juanzi Li