Saved in:
Bibliographic Details
Main Authors: Zhang, Yuqi, Miao, Yuchun, Li, Zuchao, Ding, Liang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.24519
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916768913555456
author Zhang, Yuqi
Miao, Yuchun
Li, Zuchao
Ding, Liang
author_facet Zhang, Yuqi
Miao, Yuchun
Li, Zuchao
Ding, Liang
contents We introduce AMIA, a lightweight, inference-only defense for Large Vision-Language Models (LVLMs) that (1) Automatically Masks a small set of text-irrelevant image patches to disrupt adversarial perturbations, and (2) conducts joint Intention Analysis to uncover and mitigate hidden harmful intents before response generation. Without any retraining, AMIA improves defense success rates across diverse LVLMs and jailbreak benchmarks from an average of 52.4% to 81.7%, preserves general utility with only a 2% average accuracy drop, and incurs only modest inference overhead. Ablation confirms both masking and intention analysis are essential for a robust safety-utility trade-off.
format Preprint
id arxiv_https___arxiv_org_abs_2505_24519
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AMIA: Automatic Masking and Joint Intention Analysis Makes LVLMs Robust Jailbreak Defenders
Zhang, Yuqi
Miao, Yuchun
Li, Zuchao
Ding, Liang
Computer Vision and Pattern Recognition
Computation and Language
We introduce AMIA, a lightweight, inference-only defense for Large Vision-Language Models (LVLMs) that (1) Automatically Masks a small set of text-irrelevant image patches to disrupt adversarial perturbations, and (2) conducts joint Intention Analysis to uncover and mitigate hidden harmful intents before response generation. Without any retraining, AMIA improves defense success rates across diverse LVLMs and jailbreak benchmarks from an average of 52.4% to 81.7%, preserves general utility with only a 2% average accuracy drop, and incurs only modest inference overhead. Ablation confirms both masking and intention analysis are essential for a robust safety-utility trade-off.
title AMIA: Automatic Masking and Joint Intention Analysis Makes LVLMs Robust Jailbreak Defenders
topic Computer Vision and Pattern Recognition
Computation and Language
url https://arxiv.org/abs/2505.24519