Enhanced MLLM Black-Box Jailbreaking Attacks and Defenses

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhong, Xingwei, Fok, Kar Wai, Thing, Vrizlynn L. L.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909867389747200
author Zhong, Xingwei
Fok, Kar Wai
Thing, Vrizlynn L. L.
author_facet Zhong, Xingwei
Fok, Kar Wai
Thing, Vrizlynn L. L.
contents Multimodal large language models (MLLMs) comprise of both visual and textual modalities to process vision language tasks. However, MLLMs are vulnerable to security-related issues, such as jailbreak attacks that alter the model's input to induce unauthorized or harmful responses. The incorporation of the additional visual modality introduces new dimensions to security threats. In this paper, we proposed a black-box jailbreak method via both text and image prompts to evaluate MLLMs. In particular, we designed text prompts with provocative instructions, along with image prompts that introduced mutation and multi-image capabilities. To strengthen the evaluation, we also designed a Re-attack strategy. Empirical results show that our proposed work can improve capabilities to assess the security of both open-source and closed-source MLLMs. With that, we identified gaps in existing defense methods to propose new strategies for both training-time and inference-time defense methods, and evaluated them across the new jailbreak methods. The experiment results showed that the re-designed defense methods improved protections against the jailbreak attacks.
format Preprint
id arxiv_https___arxiv_org_abs_2510_21214
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Enhanced MLLM Black-Box Jailbreaking Attacks and Defenses
Zhong, Xingwei
Fok, Kar Wai
Thing, Vrizlynn L. L.
Cryptography and Security
Multimodal large language models (MLLMs) comprise of both visual and textual modalities to process vision language tasks. However, MLLMs are vulnerable to security-related issues, such as jailbreak attacks that alter the model's input to induce unauthorized or harmful responses. The incorporation of the additional visual modality introduces new dimensions to security threats. In this paper, we proposed a black-box jailbreak method via both text and image prompts to evaluate MLLMs. In particular, we designed text prompts with provocative instructions, along with image prompts that introduced mutation and multi-image capabilities. To strengthen the evaluation, we also designed a Re-attack strategy. Empirical results show that our proposed work can improve capabilities to assess the security of both open-source and closed-source MLLMs. With that, we identified gaps in existing defense methods to propose new strategies for both training-time and inference-time defense methods, and evaluated them across the new jailbreak methods. The experiment results showed that the re-designed defense methods improved protections against the jailbreak attacks.
title Enhanced MLLM Black-Box Jailbreaking Attacks and Defenses
topic Cryptography and Security
url https://arxiv.org/abs/2510.21214