One Prompt to Verify Your Models: Black-Box Text-to-Image Models Verification via Non-Transferable Adversarial Attacks

TL;DR

This paper introduces VerifyPrompt, a method for verifying black-box text-to-image models by generating adversarial prompts that produce distinctive images, achieving over 90% accuracy in model verification tasks.

Contribution

It proposes a novel verification approach using non-transferable adversarial prompts optimized with genetic algorithms, addressing security concerns in third-party T2I services.

Findings

Achieves over 90% verification accuracy

Effective in real-world model platforms like Hugging Face

Utilizes NSGA-II for prompt optimization

Abstract

Recently, various types of Text-to-Image (T2I) models have emerged (such as DALL-E and Stable Diffusion), and showing their advantages in different aspects. Therefore, some third-party service platforms collect different model interfaces and provide cheaper API services and more flexibility in T2I model selections. However, this also raises a new security concern: Are these third-party services truly offering the models they claim? To answer this question, we first define the concept of T2I model verification, which aims to determine whether a black-box target model is identical to a given white-box reference T2I model. After that, we propose VerifyPrompt, which performs T2I model verification through a special designed verify prompt. Intuitionally, the verify prompt is an adversarial prompt for the target model without transferability for other models. It makes the target model generate a specific image while making other models produce entirely different images. Specifically, VerifyPrompt utilizes the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to optimize the cosine similarity of a prompt's text encoding, generating verify prompts. Finally, by computing the CLIP-text similarity scores between the prompts the generated images, VerifyPrompt can determine whether the target model aligns with the reference model. Experimental results demonstrate that VerifyPrompt consistently achieves over 90\% accuracy across various T2I models, confirming its effectiveness in practical model platforms (such as Hugging Face).