Efficient and Stealthy Jailbreak Attacks via Adversarial Prompt Distillation from LLMs to SLMs

📅 2025-05-26
🏛️ arXiv.org
📈 Citations: 4
Influential: 0
📄 PDF
🤖 AI Summary
Jailbreaking large language models (LLMs) incurs high computational overhead and limited practicality due to reliance on costly LLM-based inference. Method: This paper proposes the first jailbreak-capability-oriented prompt distillation framework, transferring jailbreak knowledge efficiently from LLMs to small language models (SLMs). It integrates masked language modeling, reinforcement learning, and dynamic temperature control to construct a lightweight adversarial prompt generation and distillation pipeline. Contributions/Results: Experiments demonstrate high attack success rates across multiple mainstream LLMs, strong zero-shot cross-model transferability, and enhanced attack harmfulness. The approach reduces computational cost by 73% compared to LLM-based baselines, significantly improving stealthiness and real-world deployability of jailbreak attacks.

Technology Category

Application Category

📝 Abstract
Attacks on large language models (LLMs) in jailbreaking scenarios raise many security and ethical issues. Current jailbreak attack methods face problems such as low efficiency, high computational cost, and poor cross-model adaptability and versatility, which make it difficult to cope with the rapid development of LLM and new defense strategies. Our work proposes an Adversarial Prompt Distillation, which combines masked language modeling, reinforcement learning, and dynamic temperature control through a prompt generation and distillation method. It enables small language models (SLMs) to jailbreak attacks on mainstream LLMs. The experimental results verify the superiority of the proposed method in terms of attack success rate and harm, and reflect the resource efficiency and cross-model adaptability. This research explores the feasibility of distilling the jailbreak ability of LLM to SLM, reveals the model's vulnerability, and provides a new idea for LLM security research.
Problem

Research questions and friction points this paper is trying to address.

Distilling LLM jailbreaking capabilities into smaller models
Enhancing efficiency and stealth of jailbreak attacks
Addressing deployment and reasoning constraints in LLM security
Innovation

Methods, ideas, or system contributions that make the work stand out.

Distills LLM jailbreaking into smaller models via adversarial prompt distillation
Uses masked language modeling, reinforcement learning, and dynamic temperature control
Enables efficient, stealthy attacks with high success rates and cross-model versatility
🔎 Similar Papers
No similar papers found.
X
Xiang Li
Xi’an Jiaotong-Liverpool University, The Chinese University of Hong Kong
C
Chong Zhang
Xi’an Jiaotong-Liverpool University, University of Liverpool
J
Jia Wang
Xi’an Jiaotong-Liverpool University
F
Fangyu Wu
Xi’an Jiaotong-Liverpool University
Y
Yushi Li
Xi’an Jiaotong-Liverpool University
X
Xiaobo Jin
Xi’an Jiaotong-Liverpool University