English(EN)Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control
新的LLM框架解决了多用户系统中隐私-效用权衡问题
作者PulseAugur 编辑部·[5 个来源]·
研究人员开发了新的框架和方法来解决大型语言模型(LLM)交互中的隐私问题。一种名为AIM的方法为多代理、多用户LLM环境引入了一个隐私感知内存系统,允许动态地将信息分类为私有或公共,以改善协调并保护敏感数据。另一项研究系统地分析了LLM交互中的隐私-效用权衡,提出了一种意图驱动的框架,该框架动态地清理数据以保持效用。此外,BodhiPromptShield为LLM代理管道提供了一个预推理中介层,在敏感跨度传播到系统之前检测并替换它们,从而显著减少残余暴露。
AI
arXiv:2609.18864v1 Announce Type: cross Abstract: Privacy evaluations of tool-using LLM agents often inspect a designated action, final response, or attacker report. These local proxies can miss unauthorized exposure elsewhere in a multi-step session and lack common ground truth …
arXiv:2609.14003v1 Announce Type: cross Abstract: Personal AI agents built on large language models (LLMs) are increasingly given access to a user's private data and communications in order to provide personalized assistance. This access creates a persistent privacy risk: the age…
arXiv cs.LG
TIER_1English(EN)·Zachary Johnson, Nigel Boachie Kumankumah, Somya Chatterjee, Tejas Sathyamurthi, Min Chen, Xinyi Alice Li, Xiao Wang, Emily Morgan Gelchie, Jessica Lin, Sadid A. Hasan, Sulaiman Vesal·
arXiv:2609.12320v1 Announce Type: cross Abstract: Traditional large language models (LLMs) are scoped to individual user sessions, limiting their knowledge to a single conversation and preventing them from learning user preferences that evolve over time. Existing agentic memory s…
arXiv:2609.10992v1 Announce Type: new Abstract: The integration of Large Language Models into daily tasks relies on context-rich instructions, inevitably exposing sensitive user information. Current privacy-preserving methods typically employ context-agnostic static rules, causin…
arXiv:2604.05793v2 Announce Type: replace-cross Abstract: In LLM agent pipelines, prompt privacy risk propagates beyond a single model call: raw user content enters retrieval queries, memory writes, tool arguments, OCR-derived text, and logs, and every downstream copy inherits wh…