English(EN) InferPriv: Evaluating Inferential Privacy Leakage in LLM Assistants

新基准揭示大型语言模型助手存在重大的推理隐私风险

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-27 04:00

一个名为InferPriv的新基准已被开发出来，用于评估大型语言模型（LLM）助手中的推理隐私泄露。该基准衡量LLM回复中可能隐式泄露用户敏感属性的间接线索，超越了仅评估直接泄露的现有方法。通过测试包括GPT-6.1-Sol、Sonnet-5.5和Gemini-3.8-Flash在内的五个前沿模型，研究发现，即使通过隐私感知系统提示将直接泄露降至最低，推理泄露率仍然很高（16.2-25.2%）。研究结果强调，仅评估显式隐私泄露会严重低估LLM助手带来的实际隐私风险。 AI

影响强调了对LLM助手进行更严格隐私评估的必要性，可能影响未来的模型开发和部署。

排序理由学术论文，介绍了一个用于评估LLM隐私的新基准。[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; 学术论文，介绍了一个用于评估LLM隐私的新基准。[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) · Pengrun Huang, Yiwei Fu, Luoxi Meng, Kamalika Chaudhuri · 2026-10-27 04:00

InferPriv：评估大型语言模型助手中的推理隐私泄露

arXiv:2610.34640v1 Announce Type: new Abstract: Large language model (LLM)-based personal agents are increasingly deployed to handle everyday tasks on users' behalf. These agents have access to private user information, while their tasks involve communicating with third parties; …

本文如何被排名

报道来源 [1]

InferPriv：评估大型语言模型助手中的推理隐私泄露

相关实体

相关话题