Membership Inference Attacks against Large Vision-Language Models

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Zhan, Wu, Yongtao, Chen, Yihang, Tonin, Francesco, Rocamora, Elias Abad, Cevher, Volkan
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912105758720000
author Li, Zhan
Wu, Yongtao
Chen, Yihang
Tonin, Francesco
Rocamora, Elias Abad
Cevher, Volkan
author_facet Li, Zhan
Wu, Yongtao
Chen, Yihang
Tonin, Francesco
Rocamora, Elias Abad
Cevher, Volkan
contents Large vision-language models (VLLMs) exhibit promising capabilities for processing multi-modal tasks across various application scenarios. However, their emergence also raises significant data security concerns, given the potential inclusion of sensitive information, such as private photos and medical records, in their training datasets. Detecting inappropriately used data in VLLMs remains a critical and unresolved issue, mainly due to the lack of standardized datasets and suitable methodologies. In this study, we introduce the first membership inference attack (MIA) benchmark tailored for various VLLMs to facilitate training data detection. Then, we propose a novel MIA pipeline specifically designed for token-level image detection. Lastly, we present a new metric called MaxRényi-K%, which is based on the confidence of the model output and applies to both text and image data. We believe that our work can deepen the understanding and methodology of MIAs in the context of VLLMs. Our code and datasets are available at https://github.com/LIONS-EPFL/VL-MIA.
format Preprint
id arxiv_https___arxiv_org_abs_2411_02902
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Membership Inference Attacks against Large Vision-Language Models
Li, Zhan
Wu, Yongtao
Chen, Yihang
Tonin, Francesco
Rocamora, Elias Abad
Cevher, Volkan
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Cryptography and Security
Machine Learning
Large vision-language models (VLLMs) exhibit promising capabilities for processing multi-modal tasks across various application scenarios. However, their emergence also raises significant data security concerns, given the potential inclusion of sensitive information, such as private photos and medical records, in their training datasets. Detecting inappropriately used data in VLLMs remains a critical and unresolved issue, mainly due to the lack of standardized datasets and suitable methodologies. In this study, we introduce the first membership inference attack (MIA) benchmark tailored for various VLLMs to facilitate training data detection. Then, we propose a novel MIA pipeline specifically designed for token-level image detection. Lastly, we present a new metric called MaxRényi-K%, which is based on the confidence of the model output and applies to both text and image data. We believe that our work can deepen the understanding and methodology of MIAs in the context of VLLMs. Our code and datasets are available at https://github.com/LIONS-EPFL/VL-MIA.
title Membership Inference Attacks against Large Vision-Language Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Cryptography and Security
Machine Learning
url https://arxiv.org/abs/2411.02902