基于小语言模型的 RAG 流水线上下文感知提示注入检测:nokast-secureRAG
作者:Abhishek Rai 原文链接:https://arxiv.org/html/2610.28364v1 摘要 本地检索增强生成(Retrieval-Augmented Generation, RAG)系统让强大的智能助手能够在私有数据上运行,但也继承了一类日益增长的安全风险——尤其是间接提示注入(Indirect Prompt Injection, IPI)与检索投毒,即隐藏在检索内容中...
Read on Seebug Paper (Knownsec 404)
| Source | Seebug Paper (Knownsec 404) |
|---|---|
| Fetched | 10/27/2026, 06:57 PM |
| Data | Markdown JSON |
On 10/23/2026, 05:33 PM, Seebug Paper (Knownsec 404) published "基于小语言模型的 RAG 流水线上下文感知提示注入检测:nokast-secureRAG" (type: Security news).