PulseAugur
中
实时 00:00:56
English(EN) CrackedPDFs: A Controlled Benchmark for Hidden Prompt Injection in PDFs

新的基准测试CrackedPDFs针对PDF中的隐藏提示注入

研究人员推出CrackedPDFs,这是一个新的基准测试,旨在测试大型语言模型(LLM)系统在PDF文档中嵌入的隐藏提示注入攻击。该基准测试包含超过29,000个生成的PDF,包括恶意和良性文件,用于评估检测方法。初步评估表明,一种考虑文档结构和文本的混合检测模型,其准确性很高,优于仅依赖提取文本或结构信息的模型。 AI

影响 该基准测试可能导致更强大的防御措施,以应对文档处理AI系统中复杂的提示注入攻击。

排序理由 该项目是一篇研究论文,详细介绍了一个用于评估LLM安全性的新基准测试。[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 →

新的基准测试CrackedPDFs针对PDF中的隐藏提示注入

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该项目是一篇研究论文,详细介绍了一个用于评估LLM安全性的新基准测试。[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
77 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.AI TIER_1 English(EN) · Pukaphol Thienpreecha ("Volk") ·

    CrackedPDFs:PDF中隐藏提示注入的受控基准测试

    arXiv:2607.19396v1 Announce Type: new Abstract: Document-based LLM systems often flatten a PDF before guardrails inspect it. That step can discard evidence that an instruction was never visible to the user. We introduce CrackedPDFs, a controlled benchmark for hidden prompt inject…