Why does weak-OOD help? A Further Step Towards Understanding Jailbreaking VLMs

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhou, Yuxuan, Peng, Yuzhao, Bai, Yang, Gao, Kuofeng, Zhang, Yihao, Zhang, Yechao, Chen, Xun, Yu, Tao, Dai, Tao, Xia, Shu-Tao
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911259570470912
author Zhou, Yuxuan
Peng, Yuzhao
Bai, Yang
Gao, Kuofeng
Zhang, Yihao
Zhang, Yechao
Chen, Xun
Yu, Tao
Dai, Tao
Xia, Shu-Tao
author_facet Zhou, Yuxuan
Peng, Yuzhao
Bai, Yang
Gao, Kuofeng
Zhang, Yihao
Zhang, Yechao
Chen, Xun
Yu, Tao
Dai, Tao
Xia, Shu-Tao
contents Large Vision-Language Models (VLMs) are susceptible to jailbreak attacks: researchers have developed a variety of attack strategies that can successfully bypass the safety mechanisms of VLMs. Among these approaches, jailbreak methods based on the Out-of-Distribution (OOD) strategy have garnered widespread attention due to their simplicity and effectiveness. This paper further advances the in-depth understanding of OOD-based VLM jailbreak methods. Experimental results demonstrate that jailbreak samples generated via mild OOD strategies exhibit superior performance in circumventing the safety constraints of VLMs--a phenomenon we define as ''weak-OOD''. To unravel the underlying causes of this phenomenon, this study takes SI-Attack, a typical OOD-based jailbreak method, as the research object. We attribute this phenomenon to a trade-off between two dominant factors: input intent perception and model refusal triggering. The inconsistency in how these two factors respond to OOD manipulations gives rise to this phenomenon. Furthermore, we provide a theoretical argument for the inevitability of such inconsistency from the perspective of discrepancies between model pre-training and alignment processes. Building on the above insights, we draw inspiration from optical character recognition (OCR) capability enhancement--a core task in the pre-training phase of mainstream VLMs. Leveraging this capability, we design a simple yet highly effective VLM jailbreak method, whose performance outperforms that of SOTA baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2511_08367
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Why does weak-OOD help? A Further Step Towards Understanding Jailbreaking VLMs
Zhou, Yuxuan
Peng, Yuzhao
Bai, Yang
Gao, Kuofeng
Zhang, Yihao
Zhang, Yechao
Chen, Xun
Yu, Tao
Dai, Tao
Xia, Shu-Tao
Cryptography and Security
Large Vision-Language Models (VLMs) are susceptible to jailbreak attacks: researchers have developed a variety of attack strategies that can successfully bypass the safety mechanisms of VLMs. Among these approaches, jailbreak methods based on the Out-of-Distribution (OOD) strategy have garnered widespread attention due to their simplicity and effectiveness. This paper further advances the in-depth understanding of OOD-based VLM jailbreak methods. Experimental results demonstrate that jailbreak samples generated via mild OOD strategies exhibit superior performance in circumventing the safety constraints of VLMs--a phenomenon we define as ''weak-OOD''. To unravel the underlying causes of this phenomenon, this study takes SI-Attack, a typical OOD-based jailbreak method, as the research object. We attribute this phenomenon to a trade-off between two dominant factors: input intent perception and model refusal triggering. The inconsistency in how these two factors respond to OOD manipulations gives rise to this phenomenon. Furthermore, we provide a theoretical argument for the inevitability of such inconsistency from the perspective of discrepancies between model pre-training and alignment processes. Building on the above insights, we draw inspiration from optical character recognition (OCR) capability enhancement--a core task in the pre-training phase of mainstream VLMs. Leveraging this capability, we design a simple yet highly effective VLM jailbreak method, whose performance outperforms that of SOTA baselines.
title Why does weak-OOD help? A Further Step Towards Understanding Jailbreaking VLMs
topic Cryptography and Security
url https://arxiv.org/abs/2511.08367