English(EN) Detecting and Mitigating Query-Based System Prompt Leakage Attacks

新的AdvGuard框架应对LLM提示泄露攻击

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-28 04:00

研究人员开发了AdvGuard，一个旨在检测和缓解针对大型语言模型（LLM）的提示泄露攻击的新型框架。这些攻击旨在提取指导LLM行为的机密系统提示。AdvGuard利用图卷积网络（GCN）来模拟用户查询之间的关系，识别跨多次交互出现的恶意模式。该框架还包括一个用于即时保护的提示级检测器，并在评估中表现出高准确率，显著降低了已知和未知攻击家族的成功率。 AI

影响通过提供强大的提示泄露防御能力来增强LLM安全性，这对于保护专有信息和确保模型完整性至关重要。

排序理由详细介绍LLM安全新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 14 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍LLM安全新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, safety; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Tushar Nayan, Pratik Poudel, Qian Lou, Yuede Ji, Jason Liu, Ruimin Sun · 2026-10-28 04:00

检测和缓解基于查询的系统提示泄露攻击

arXiv:2610.36295v1 Announce Type: cross Abstract: Large Language Models (LLMs) rely on proprietary system prompts to enforce task directives, safety policies, and access controls. These prompts are vulnerable to query-based leakage attacks, where black-box adversaries extract con…

本文如何被排名

报道来源 [1]

检测和缓解基于查询的系统提示泄露攻击

相关实体

相关话题