Buchumschlag

Same Model, Different Weakness: How Language and Modality Reshape the Jailbreak Attack Surface in Frontier MLLMs

Fuente: arXiv

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Ford, Casey, Van Doren, Madison, Jin, Sicheng, Dix, Emily
Format:	Preprint
Veröffentlicht:	2026
Schlagworte:	Computation and Language
Online-Zugang:	Acceder al recurso
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Ähnliche Einträge

Alignment Drift in Multimodal LLMs: A Two-Phase, Longitudinal Evaluation of Harm Across Eight Model Releases
von: Ford, Casey, et al.
Veröffentlicht: (2026)

Red Teaming Multimodal Language Models: Evaluating Harm Across Prompt Modalities and Models
von: Van Doren, Madison, et al.
Veröffentlicht: (2025)

"Be My Cheese?": Cultural Nuance Benchmarking for Machine Translation in Multilingual LLMs
von: Van Doren, Madison, et al.
Veröffentlicht: (2026)

"Be My Cheese?": Assessing Cultural Nuance in Multilingual LLM Translations
von: Van Doren, Madison, et al.
Veröffentlicht: (2025)

AttackEval: How to Evaluate the Effectiveness of Jailbreak Attacking on Large Language Models
von: Shu, Dong, et al.
Veröffentlicht: (2024)

Cross-Modal Obfuscation for Jailbreak Attacks on Large Vision-Language Models
von: Jiang, Lei, et al.
Veröffentlicht: (2025)

Weak-to-Strong Jailbreaking on Large Language Models
von: Zhao, Xuandong, et al.
Veröffentlicht: (2024)

Emerging Vulnerabilities in Frontier Models: Multi-Turn Jailbreak Attacks
von: Gibbs, Tom, et al.
Veröffentlicht: (2024)

Visual Contextual Attack: Jailbreaking MLLMs with Image-Driven Context Injection
von: Miao, Ziqi, et al.
Veröffentlicht: (2025)

Jailbreaking Large Language Models with Morality Attacks
von: Su, Ying, et al.
Veröffentlicht: (2026)

The Same But Different: Structural Similarities and Differences in Multilingual Language Modeling
von: Zhang, Ruochen, et al.
Veröffentlicht: (2024)

Jailbreaking Attack against Multimodal Large Language Model
von: Niu, Zhenxing, et al.
Veröffentlicht: (2024)

Exploring How Generative MLLMs Perceive More Than CLIP with the Same Vision Encoder
von: Li, Siting, et al.
Veröffentlicht: (2024)

Intrinsic Model Weaknesses: How Priming Attacks Unveil Vulnerabilities in Large Language Models
von: Huang, Yuyi, et al.
Veröffentlicht: (2025)

Structured Semantic Cloaking for Jailbreak Attacks on Large Language Models
von: Sun, Xiaobing, et al.
Veröffentlicht: (2026)

JailbreakLens: Visual Analysis of Jailbreak Attacks Against Large Language Models
von: Feng, Yingchaojie, et al.
Veröffentlicht: (2024)

TAO-Attack: Toward Advanced Optimization-Based Jailbreak Attacks for Large Language Models
von: Xu, Zhi, et al.
Veröffentlicht: (2026)

Response Attack: Exploiting Contextual Priming to Jailbreak Large Language Models
von: Miao, Ziqi, et al.
Veröffentlicht: (2025)

From LLMs to MLLMs: Exploring the Landscape of Multimodal Jailbreaking
von: Wang, Siyuan, et al.
Veröffentlicht: (2024)

Knowledge-to-Jailbreak: Investigating Knowledge-driven Jailbreaking Attacks for Large Language Models
von: Tu, Shangqing, et al.
Veröffentlicht: (2024)

Medical MLLM is Vulnerable: Cross-Modality Jailbreak and Mismatched Attacks on Medical Multimodal Large Language Models
von: Huang, Xijie, et al.
Veröffentlicht: (2024)

Defending Large Language Models Against Jailbreaking Attacks Through Goal Prioritization
von: Zhang, Zhexin, et al.
Veröffentlicht: (2023)

Defending Large Language Models against Jailbreak Attacks via Semantic Smoothing
von: Ji, Jiabao, et al.
Veröffentlicht: (2024)

We're Different, We're the Same: Creative Homogeneity Across LLMs
von: Wenger, Emily, et al.
Veröffentlicht: (2025)

Distract Large Language Models for Automatic Jailbreak Attack
von: Xiao, Zeguan, et al.
Veröffentlicht: (2024)

PandaGuard: Systematic Evaluation of LLM Safety against Jailbreaking Attacks
von: Shen, Guobin, et al.
Veröffentlicht: (2025)

Harnessing Task Overload for Scalable Jailbreak Attacks on Large Language Models
von: Dong, Yiting, et al.
Veröffentlicht: (2024)

Unified Defense for Large Language Models against Jailbreak and Fine-Tuning Attacks in Education
von: Yi, Xin, et al.
Veröffentlicht: (2025)

Jailbreaking Leaves a Trace: Understanding and Detecting Jailbreak Attacks from Internal Representations of Large Language Models
von: Kadali, Sri Durga Sai Sowmya, et al.
Veröffentlicht: (2026)

Jailbreaking Frontier Foundation Models Through Intention Deception
von: Wang, Xinhe, et al.
Veröffentlicht: (2026)

Same Task, Different Circuits: Disentangling Modality-Specific Mechanisms in VLMs
von: Nikankin, Yaniv, et al.
Veröffentlicht: (2025)

JailBreakV: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks
von: Luo, Weidi, et al.
Veröffentlicht: (2024)

StructuralSleight: Automated Jailbreak Attacks on Large Language Models Utilizing Uncommon Text-Organization Structures
von: Li, Bangxin, et al.
Veröffentlicht: (2024)

LLM-Virus: Evolutionary Jailbreak Attack on Large Language Models
von: Yu, Miao, et al.
Veröffentlicht: (2024)

Quantisation Reshapes the Metacognitive Geometry of Language Models
von: Cacioli, Jon-Paul
Veröffentlicht: (2026)

Enhancing Jailbreak Attacks with Diversity Guidance
von: Zhang, Xu, et al.
Veröffentlicht: (2024)

Stealthy Jailbreak Attacks on Large Language Models via Benign Data Mirroring
von: Mu, Honglin, et al.
Veröffentlicht: (2024)

Round Trip Translation Defence against Large Language Model Jailbreaking Attacks
von: Yung, Canaan, et al.
Veröffentlicht: (2024)

UniGuard: Towards Universal Safety Guardrails for Jailbreak Attacks on Multimodal Large Language Models
von: Oh, Sejoon, et al.
Veröffentlicht: (2024)

All in How You Ask for It: Simple Black-Box Method for Jailbreak Attacks
von: Takemoto, Kazuhiro
Veröffentlicht: (2024)