English(EN) Auditing and Mitigating Privacy Leakage in Cloud-Edge Collaborative Decoding

新框架审计和缓解云边大语言模型协作中的隐私泄露

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-01 04:00

研究人员开发了一个新框架，用于审计和缓解云边协同解码系统中的隐私泄露。这些系统使用边缘设备上的小型语言模型处理私有数据，并将其预测与云端托管的大型语言模型融合。所提出的名为CoVeil的方法动态优化传输信号，将数据泄露减少高达87.2%，同时保持协同质量，这在构建的问答数据集上得到了证明。 AI

影响这项研究可能导致更安全地部署大语言模型来处理敏感数据，从而在注重隐私的应用中得到更广泛的应用。

排序理由这是一篇发表在arXiv上的研究论文，详细介绍了一个新的框架和方法，用于解决特定AI系统架构中的隐私问题。[lever_c_demoted from research: ic=1 ai=1.0]

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: 25 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 这是一篇发表在arXiv上的研究论文，详细介绍了一个新的框架和方法，用于解决特定AI系统架构中的隐私问题。[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。

arXiv cs.AI TIER_1 English(EN) · Kejia Zhang, Tianyuan Zou, Zixuan GU, Yang Liu · 2026-09-01 04:00

云边协同解码中的隐私泄露审计与缓解

arXiv:2608.29111v1 Announce Type: cross Abstract: Applications such as personalized assistance and proprietary document analysis require large language models (LLMs) to generate outputs from private data. Yet powerful LLMs typically cannot be deployed on the resource-constrained …