paper-with-me

Papers

The Phantom Menace: Unmasking Privacy Leakages in Vision-Language Models

2024-08-02 · Simone Caldarella, Massimiliano Mancini, Elisa Ricci, Rahaf Aljundi

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.

📄 PDF Abstract BibTeX arXiv:2408.01228

Code (0)

등록된 구현이 없습니다.

Tasks

Image Captioning

Similar Papers 제목 키워드 기반

Membership Inference Attack Susceptibility of Clinical Language Models

2021-04-16 · Abhyuday Jagannatha, Bhanu Pratap Singh Rawat, Hong Yu

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 Attack

CenterYou: A cloud-based Approach to Simplify Android Privacy Management

2020-08-31 · Seyedmostafa Safavi, Zarina Shukur

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 MakingManagement

Safe-LLaVA: A Privacy-Preserving Vision-Language Dataset and Benchmark for Biometric Safety

2025-08-29 · Younggun Kim, Sirnam Swetha, Fazil Kagdi, Mubarak Shah arxiv

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

2022-08-23 · Zheng Li, Yiyong Liu, Xinlei He, Ning Yu 외

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?

2018-11-13 · Manaar Alam, Debdeep Mukhopadhyay

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