On the Multi-modal Vulnerability of Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Dingcheng, Bai, Yang, Jia, Xiaojun, Liu, Yang, Cao, Xiaochun, Yu, Wenjian
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915088355557376
author Yang, Dingcheng
Bai, Yang
Jia, Xiaojun
Liu, Yang
Cao, Xiaochun
Yu, Wenjian
author_facet Yang, Dingcheng
Bai, Yang
Jia, Xiaojun
Liu, Yang
Cao, Xiaochun
Yu, Wenjian
contents Diffusion models have been widely deployed in various image generation tasks, demonstrating an extraordinary connection between image and text modalities. Although prior studies have explored the vulnerability of diffusion models from the perspectives of text and image modalities separately, the current research landscape has not yet thoroughly investigated the vulnerabilities that arise from the integration of multiple modalities, specifically through the joint analysis of textual and visual features. In this paper, we are the first to visualize both text and image feature space embedded by diffusion models and observe a significant difference. The prompts are embedded chaotically in the text feature space, while in the image feature space they are clustered according to their subjects. These fascinating findings may underscore a potential misalignment in robustness between the two modalities that exists within diffusion models. Based on this observation, we propose MMP-Attack, which leverages multi-modal priors (MMP) to manipulate the generation results of diffusion models by appending a specific suffix to the original prompt. Specifically, our goal is to induce diffusion models to generate a specific object while simultaneously eliminating the original object. Our MMP-Attack shows a notable advantage over existing studies with superior manipulation capability and efficiency. Our code is publicly available at \url{https://github.com/ydc123/MMP-Attack}.
format Preprint
id arxiv_https___arxiv_org_abs_2402_01369
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle On the Multi-modal Vulnerability of Diffusion Models
Yang, Dingcheng
Bai, Yang
Jia, Xiaojun
Liu, Yang
Cao, Xiaochun
Yu, Wenjian
Machine Learning
Cryptography and Security
Computer Vision and Pattern Recognition
Diffusion models have been widely deployed in various image generation tasks, demonstrating an extraordinary connection between image and text modalities. Although prior studies have explored the vulnerability of diffusion models from the perspectives of text and image modalities separately, the current research landscape has not yet thoroughly investigated the vulnerabilities that arise from the integration of multiple modalities, specifically through the joint analysis of textual and visual features. In this paper, we are the first to visualize both text and image feature space embedded by diffusion models and observe a significant difference. The prompts are embedded chaotically in the text feature space, while in the image feature space they are clustered according to their subjects. These fascinating findings may underscore a potential misalignment in robustness between the two modalities that exists within diffusion models. Based on this observation, we propose MMP-Attack, which leverages multi-modal priors (MMP) to manipulate the generation results of diffusion models by appending a specific suffix to the original prompt. Specifically, our goal is to induce diffusion models to generate a specific object while simultaneously eliminating the original object. Our MMP-Attack shows a notable advantage over existing studies with superior manipulation capability and efficiency. Our code is publicly available at \url{https://github.com/ydc123/MMP-Attack}.
title On the Multi-modal Vulnerability of Diffusion Models
topic Machine Learning
Cryptography and Security
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2402.01369