Saved in:
Bibliographic Details
Main Authors: Liu, Sheng, Sheng, Qiang, Wang, Danding, Li, Yang, Yang, Guang, Cao, Juan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.20038
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909770171023360
author Liu, Sheng
Sheng, Qiang
Wang, Danding
Li, Yang
Yang, Guang
Cao, Juan
author_facet Liu, Sheng
Sheng, Qiang
Wang, Danding
Li, Yang
Yang, Guang
Cao, Juan
contents Despite advances in improving large language model (LLM) to refuse to answer malicious instructions, widely used LLMs remain vulnerable to jailbreak attacks where attackers generate instructions with distributions differing from safety alignment corpora. New attacks expose LLMs' inability to recognize unseen malicious instructions, highlighting a critical distributional mismatch between training data and real-world attacks that forces developers into reactive patching cycles. To tackle this challenge, we propose IMAGINE, a synthesis framework that leverages embedding space distribution analysis to generate jailbreak-like instructions. This approach effectively fills the distributional gap between authentic jailbreak patterns and safety alignment corpora. IMAGINE follows an iterative optimization process that dynamically evolves text generation distributions across iterations, thereby augmenting the coverage of safety alignment data distributions through synthesized data examples. Based on the safety-aligned corpus enhanced through IMAGINE, our framework demonstrates significant decreases in attack success rate on Qwen2.5, Llama3.1, and Llama3.2 without compromising their utility.
format Preprint
id arxiv_https___arxiv_org_abs_2508_20038
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Forewarned is Forearmed: Pre-Synthesizing Jailbreak-like Instructions to Enhance LLM Safety Guardrail to Potential Attacks
Liu, Sheng
Sheng, Qiang
Wang, Danding
Li, Yang
Yang, Guang
Cao, Juan
Computation and Language
Despite advances in improving large language model (LLM) to refuse to answer malicious instructions, widely used LLMs remain vulnerable to jailbreak attacks where attackers generate instructions with distributions differing from safety alignment corpora. New attacks expose LLMs' inability to recognize unseen malicious instructions, highlighting a critical distributional mismatch between training data and real-world attacks that forces developers into reactive patching cycles. To tackle this challenge, we propose IMAGINE, a synthesis framework that leverages embedding space distribution analysis to generate jailbreak-like instructions. This approach effectively fills the distributional gap between authentic jailbreak patterns and safety alignment corpora. IMAGINE follows an iterative optimization process that dynamically evolves text generation distributions across iterations, thereby augmenting the coverage of safety alignment data distributions through synthesized data examples. Based on the safety-aligned corpus enhanced through IMAGINE, our framework demonstrates significant decreases in attack success rate on Qwen2.5, Llama3.1, and Llama3.2 without compromising their utility.
title Forewarned is Forearmed: Pre-Synthesizing Jailbreak-like Instructions to Enhance LLM Safety Guardrail to Potential Attacks
topic Computation and Language
url https://arxiv.org/abs/2508.20038