Text Prompt Injection of Vision Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Zhu, Ruizhe
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914087533805568
author Zhu, Ruizhe
author_facet Zhu, Ruizhe
contents The widespread application of large vision language models has significantly raised safety concerns. In this project, we investigate text prompt injection, a simple yet effective method to mislead these models. We developed an algorithm for this type of attack and demonstrated its effectiveness and efficiency through experiments. Compared to other attack methods, our approach is particularly effective for large models without high demand for computational resources.
format Preprint
id arxiv_https___arxiv_org_abs_2510_09849
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Text Prompt Injection of Vision Language Models
Zhu, Ruizhe
Computation and Language
Computer Vision and Pattern Recognition
The widespread application of large vision language models has significantly raised safety concerns. In this project, we investigate text prompt injection, a simple yet effective method to mislead these models. We developed an algorithm for this type of attack and demonstrated its effectiveness and efficiency through experiments. Compared to other attack methods, our approach is particularly effective for large models without high demand for computational resources.
title Text Prompt Injection of Vision Language Models
topic Computation and Language
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.09849