paperSeptember 2026Unreviewed

PriMobiBench: Characterizing Visual Privacy Leakage in VLM-Driven Mobile GUI Agents

Qihang Cen, Tianshuo Cong, Da Song, Xinlei He, Jiaxing Song, Ke Xu, Qi Li

Abstract

Mobile GUI agents increasingly rely on Vision-Language Models (VLMs) to automate smartphone tasks by interpreting screenshot streams. However, this design introduces serious and underexplored privacy risks, including direct leakage of sensitive on-screen information and unintended user profiling. The absence of standardized benchmarks makes it difficult to quantify these risks in realistic mobile agent workflows. To address this gap, we propose PriMobiBench, the first benchmark for systematicall

Framework mappings

OWASP Top 10 for LLM Applications

LLM02Sensitive Information Disclosure

MITRE ATLAS

AML.T0024.000Infer Training Data Membership

Suggested from the entry's categories.

Cite

@misc{cen2026primobibench,
  title = {{PriMobiBench: Characterizing Visual Privacy Leakage in VLM-Driven Mobile GUI Agents}},
  author = {Qihang Cen and Tianshuo Cong and Da Song and Xinlei He and Jiaxing Song and Ke Xu and Qi Li},
  year = {2026},
  month = sep,
  eprint = {2609.13873},
  archivePrefix = {arXiv},
  doi = {10.1145/3830454.3846776},
  url = {https://arxiv.org/abs/2609.13873}
}