Who Can See Through You? Adversarial Shielding Against VLM-Based Attribute Inference Attacks

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Fan, Yucheng, Chen, Jiawei, Tian, Yu, Yin, Zhaoxia
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909972351156224
author Fan, Yucheng
Chen, Jiawei
Tian, Yu
Yin, Zhaoxia
author_facet Fan, Yucheng
Chen, Jiawei
Tian, Yu
Yin, Zhaoxia
contents As vision-language models (VLMs) become widely adopted, VLM-based attribute inference attacks have emerged as a serious privacy concern, enabling adversaries to infer private attributes from images shared on social media. This escalating threat calls for dedicated protection methods to safeguard user privacy. However, existing methods often degrade the visual quality of images or interfere with vision-based functions on social media, thereby failing to achieve a desirable balance between privacy protection and user experience. To address this challenge, we propose a novel protection method that jointly optimizes privacy suppression and utility preservation under a visual consistency constraint. While our method is conceptually effective, fair comparisons between methods remain challenging due to the lack of publicly available evaluation datasets. To fill this gap, we introduce VPI-COCO, a publicly available benchmark comprising 522 images with hierarchically structured privacy questions and corresponding non-private counterparts, enabling fine-grained and joint evaluation of protection methods in terms of privacy preservation and user experience. Building upon this benchmark, experiments on multiple VLMs demonstrate that our method effectively reduces PAR below 25%, keeps NPAR above 88%, maintains high visual consistency, and generalizes well to unseen and paraphrased privacy questions, demonstrating its strong practical applicability for real-world VLM deployments.
format Preprint
id arxiv_https___arxiv_org_abs_2512_18264
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Who Can See Through You? Adversarial Shielding Against VLM-Based Attribute Inference Attacks
Fan, Yucheng
Chen, Jiawei
Tian, Yu
Yin, Zhaoxia
Computer Vision and Pattern Recognition
Artificial Intelligence
Cryptography and Security
As vision-language models (VLMs) become widely adopted, VLM-based attribute inference attacks have emerged as a serious privacy concern, enabling adversaries to infer private attributes from images shared on social media. This escalating threat calls for dedicated protection methods to safeguard user privacy. However, existing methods often degrade the visual quality of images or interfere with vision-based functions on social media, thereby failing to achieve a desirable balance between privacy protection and user experience. To address this challenge, we propose a novel protection method that jointly optimizes privacy suppression and utility preservation under a visual consistency constraint. While our method is conceptually effective, fair comparisons between methods remain challenging due to the lack of publicly available evaluation datasets. To fill this gap, we introduce VPI-COCO, a publicly available benchmark comprising 522 images with hierarchically structured privacy questions and corresponding non-private counterparts, enabling fine-grained and joint evaluation of protection methods in terms of privacy preservation and user experience. Building upon this benchmark, experiments on multiple VLMs demonstrate that our method effectively reduces PAR below 25%, keeps NPAR above 88%, maintains high visual consistency, and generalizes well to unseen and paraphrased privacy questions, demonstrating its strong practical applicability for real-world VLM deployments.
title Who Can See Through You? Adversarial Shielding Against VLM-Based Attribute Inference Attacks
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Cryptography and Security
url https://arxiv.org/abs/2512.18264