cs.AI updates on arXiv.org 前天 12:27
LeakSealer: A Semisupervised Defense for LLMs Against Prompt Injection and Leakage Attacks
index_new5.html
../../../zaker_core/zaker_tpl_static/wap/tpl_guoji1.html

 

本文提出了一种名为LeakSealer的模型,用于分析LLM系统历史交互数据,并针对安全威胁提供防御机制,通过实证测试在静态和动态场景下均表现出色。

arXiv:2508.00602v1 Announce Type: cross Abstract: The generalization capabilities of Large Language Models (LLMs) have led to their widespread deployment across various applications. However, this increased adoption has introduced several security threats, notably in the forms of jailbreaking and data leakage attacks. Additionally, Retrieval Augmented Generation (RAG), while enhancing context-awareness in LLM responses, has inadvertently introduced vulnerabilities that can result in the leakage of sensitive information. Our contributions are twofold. First, we introduce a methodology to analyze historical interaction data from an LLM system, enabling the generation of usage maps categorized by topics (including adversarial interactions). This approach further provides forensic insights for tracking the evolution of jailbreaking attack patterns. Second, we propose LeakSealer, a model-agnostic framework that combines static analysis for forensic insights with dynamic defenses in a Human-In-The-Loop (HITL) pipeline. This technique identifies topic groups and detects anomalous patterns, allowing for proactive defense mechanisms. We empirically evaluate LeakSealer under two scenarios: (1) jailbreak attempts, employing a public benchmark dataset, and (2) PII leakage, supported by a curated dataset of labeled LLM interactions. In the static setting, LeakSealer achieves the highest precision and recall on the ToxicChat dataset when identifying prompt injection. In the dynamic setting, PII leakage detection achieves an AUPRC of $0.97$, significantly outperforming baselines such as Llama Guard.

Fish AI Reader

Fish AI Reader

AI辅助创作,多种专业模板,深度分析,高质量内容生成。从观点提取到深度思考,FishAI为您提供全方位的创作支持。新版本引入自定义参数,让您的创作更加个性化和精准。

FishAI

FishAI

鱼阅,AI 时代的下一个智能信息助手,助你摆脱信息焦虑

联系邮箱 441953276@qq.com

相关标签

LLM安全 LeakSealer模型 数据泄露防御
相关文章