The Phantom Menace: Unmasking Privacy Leakages in Vision-Language Models
Vision-Language Models (VLMs) combine visual and textual understanding, rendering them well-suited for diverse tasks like generating image captions and answering visual questions across various domains. However, these capabilities are built upon training on large amount of uncurated data crawled from the web. The latter may include sensitive information that VLMs could memorize and leak, raising significant privacy concerns. In this paper, we assess whether these vulnerabilities exist, focusing on identity leakage. Our study leads to three key findings: (i) VLMs leak identity information, even when the vision-language alignment and the fine-tuning use anonymized data; (ii) context has little influence on identity leakage; (iii) simple, widely used anonymization techniques, like blurring, are not sufficient to address the problem. These findings underscore the urgent need for robust privacy protection strategies when deploying VLMs. Ethical awareness and responsible development practices are essential to mitigate these risks.
Code (0)
등록된 구현이 없습니다.
Tasks
Image CaptioningSimilar Papers 제목 키워드 기반
Membership Inference Attack Susceptibility of Clinical Language Models
Deep Neural Network (DNN) models have been shown to have high empirical privacy leakages. Clinical language models (CLMs) trained on clinical data have been used to improve performance in biomedical natural language proc…
Inference AttackMembership Inference AttackCenterYou: A cloud-based Approach to Simplify Android Privacy Management
With mobile applications and associated services becoming increasingly popular, concerns are being raised about private data leakages have raised. Previous solutions to this well-known set of problems have approached it …
Decision MakingManagementSafe-LLaVA: A Privacy-Preserving Vision-Language Dataset and Benchmark for Biometric Safety
Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities in vision-language tasks. However, these models often infer and reveal sensitive biometric attributes such as race, gender, age, body wei…
Auditing Membership Leakages of Multi-Exit Networks
Relying on the fact that not all inputs require the same amount of computation to yield a confident prediction, multi-exit networks are gaining attention as a prominent approach for pushing the limits of efficient deploy…
How Secure are Deep Learning Algorithms from Side-Channel based Reverse Engineering?
Deep Learning algorithms have recently become the de-facto paradigm for various prediction problems, which include many privacy-preserving applications like online medical image analysis. Presumably, the privacy of data …
Deep LearningMedical Image AnalysisPrivacy PreservingTwo-sample testing