PulseAugur
中
实时 10:42:51
English(EN) LeakageBench: Document-Level Leakage Risk for Redacting Personally Identifiable Information in Document Images

新的LeakageBench基准揭示文档脱敏中持续存在的个人身份信息(PII)风险

一个名为LeakageBench的新基准已被开发出来,用于评估文档图像中个人身份信息(PII)泄露的风险。该基准包含500张带有超过11,000个GDPR对齐的PII注释的文档图像,评估了包括OCR管道和视觉语言模型在内的各种脱敏方法。虽然像Code Interpreter这样的工具在与GPT-5.5等模型结合使用时可以提高PII的定位精度,但该基准表明,页面级别的泄露风险仍然存在,凸显了对更强大的PII脱敏技术的需求。 AI

影响 强调了AI驱动的PII脱敏方面持续存在的挑战,表明当前的工具和模型在全面的文档级安全性方面仍有不足。

排序理由 该集群描述了在arXiv上发布的一个新的学术基准和研究论文。

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

新的LeakageBench基准揭示文档脱敏中持续存在的个人身份信息(PII)风险

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
Research
该集群描述了在arXiv上发布的一个新的学术基准和研究论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
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
47 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 [2]

  1. arXiv cs.CL TIER_1 English(EN) · Vishnu Prasad Vijaya Kumar, Santhosh Venkatesh, Ivan P. Yamshchikov ·

    LeakageBench:文档图像中用于Personally Identifiable Information(PII)红色的文档级泄露风险

    arXiv:2609.02207v1 Announce Type: cross Abstract: Real-world personally identifiable information (PII) redaction often operates on document images---scans, screenshots, and PDF renderings---where OCR errors, layout structure, and visual noise determine whether sensitive informati…

  2. Hugging Face Daily Papers TIER_1 English(EN) ·

    LeakageBench:文档图像中用于编辑个人身份信息文档级别的泄露风险

    Real-world personally identifiable information (PII) redaction often operates on document images---scans, screenshots, and PDF renderings---where OCR errors, layout structure, and visual noise determine whether sensitive information is actually removed. Existing PII benchmarks ar…