Buchumschlag

Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their Defenses

Fuente: arXiv

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Zheng, Xiaosen, Pang, Tianyu, Du, Chao, Liu, Qian, Jiang, Jing, Lin, Min
Format:	Preprint
Veröffentlicht:	2024
Schlagworte:	Computation and Language Artificial Intelligence Cryptography and Security Machine Learning
Online-Zugang:	Acceder al recurso
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Ähnliche Einträge

Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates
von: Zheng, Xiaosen, et al.
Veröffentlicht: (2024)

Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast
von: Gu, Xiangming, et al.
Veröffentlicht: (2024)

Improved Techniques for Optimization-Based Jailbreaking on Large Language Models
von: Jia, Xiaojun, et al.
Veröffentlicht: (2024)

Improving Your Model Ranking on Chatbot Arena by Vote Rigging
von: Min, Rui, et al.
Veröffentlicht: (2025)

Imperceptible Jailbreaking against Large Language Models
von: Gao, Kuofeng, et al.
Veröffentlicht: (2025)

Denial-of-Service Poisoning Attacks against Large Language Models
von: Gao, Kuofeng, et al.
Veröffentlicht: (2024)

Retrieval-Augmented Defense: Adaptive and Controllable Jailbreak Prevention for Large Language Models
von: Yang, Guangyu, et al.
Veröffentlicht: (2025)

Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrations
von: Wei, Zeming, et al.
Veröffentlicht: (2023)

Test-Time Backdoor Attacks on Multimodal Large Language Models
von: Lu, Dong, et al.
Veröffentlicht: (2024)

Sparse Tokens Suffice: Jailbreaking Audio Language Models via Token-Aware Gradient Optimization
von: Fang, Zheng, et al.
Veröffentlicht: (2026)

GenBreak: Red Teaming Text-to-Image Generators Using Large Language Models
von: Wang, Zilong, et al.
Veröffentlicht: (2025)

Self-Jailbreaking: Language Models Can Reason Themselves Out of Safety Alignment After Benign Reasoning Training
von: Yong, Zheng-Xin, et al.
Veröffentlicht: (2025)

ForgeDAN: An Evolutionary Framework for Jailbreaking Aligned Large Language Models
von: Cheng, Siyang, et al.
Veröffentlicht: (2025)

Lifelong Safety Alignment for Language Models
von: Wang, Haoyu, et al.
Veröffentlicht: (2025)

BitBypass: A New Direction in Jailbreaking Aligned Large Language Models with Bitstream Camouflage
von: Nakka, Kalyan, et al.
Veröffentlicht: (2025)

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

CCJA: Context-Coherent Jailbreak Attack for Aligned Large Language Models
von: Zhou, Guanghao, et al.
Veröffentlicht: (2025)

QueryAttack: Jailbreaking Aligned Large Language Models Using Structured Non-natural Query Language
von: Zou, Qingsong, et al.
Veröffentlicht: (2025)

SecurityLingua: Efficient Defense of LLM Jailbreak Attacks via Security-Aware Prompt Compression
von: Li, Yucheng, et al.
Veröffentlicht: (2025)

AdaSteer: Your Aligned LLM is Inherently an Adaptive Jailbreak Defender
von: Zhao, Weixiang, et al.
Veröffentlicht: (2025)

Towards Safe AI Clinicians: A Comprehensive Study on Large Language Model Jailbreaking in Healthcare
von: Zhang, Hang, et al.
Veröffentlicht: (2025)

Jailbreak Attacks and Defenses Against Large Language Models: A Survey
von: Yi, Sibo, et al.
Veröffentlicht: (2024)

SafeSteer: Adaptive Subspace Steering for Efficient Jailbreak Defense in Vision-Language Models
von: Zeng, Xiyu, et al.
Veröffentlicht: (2025)

Preventing Jailbreak Prompts as Malicious Tools for Cybercriminals: A Cyber Defense Perspective
von: Tshimula, Jean Marie, et al.
Veröffentlicht: (2024)

AutoDefense: Multi-Agent LLM Defense against Jailbreak Attacks
von: Zeng, Yifan, et al.
Veröffentlicht: (2024)

JailbreakEval: An Integrated Toolkit for Evaluating Jailbreak Attempts Against Large Language Models
von: Ran, Delong, et al.
Veröffentlicht: (2024)

Testing the Limits of Jailbreaking Defenses with the Purple Problem
von: Kim, Taeyoun, et al.
Veröffentlicht: (2024)

How Jailbreak Defenses Work and Ensemble? A Mechanistic Investigation
von: Long, Zhuohang, et al.
Veröffentlicht: (2025)

Privacy-Preserving Instructions for Aligning Large Language Models
von: Yu, Da, et al.
Veröffentlicht: (2024)

Test-Time Immunization: A Universal Defense Framework Against Jailbreaks for (Multimodal) Large Language Models
von: Yu, Yongcan, et al.
Veröffentlicht: (2025)

You Can't Eat Your Cake and Have It Too: The Performance Degradation of LLMs with Jailbreak Defense
von: Mai, Wuyuao, 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)

EvoDefense: Co-Evolving Black-Box Defense with Large Language Models
von: Li, Yu, et al.
Veröffentlicht: (2026)

Don't Listen To Me: Understanding and Exploring Jailbreak Prompts of Large Language Models
von: Yu, Zhiyuan, et al.
Veröffentlicht: (2024)

InfoFlood: Jailbreaking Large Language Models with Information Overload
von: Yadav, Advait, et al.
Veröffentlicht: (2025)

from Benign import Toxic: Jailbreaking the Language Model via Adversarial Metaphors
von: Yan, Yu, et al.
Veröffentlicht: (2025)

Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned Concepts
von: Gao, Hongcheng, et al.
Veröffentlicht: (2024)

FlexLLM: Exploring LLM Customization for Moving Target Defense on Black-Box LLMs Against Jailbreak Attacks
von: Chen, Bocheng, et al.
Veröffentlicht: (2024)

Layer-Level Self-Exposure and Patch: Affirmative Token Mitigation for Jailbreak Attack Defense
von: Ouyang, Yang, et al.
Veröffentlicht: (2025)

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