English(EN) AfterDeleteBench: Auditing Post-Deletion Information Persistence in LLM Agents

新基准揭示 LLM 代理会保留已删除信息

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-23 04:00

一个名为 AfterDeleteBench 的新基准已被开发出来，用于审计 LLM 代理在原始数据被删除后信息的持久性。该基准发现，包括 Qwen2.5-7B-Instruct、Qwen、Mistral 和 Phi-3 在内的 LLM 代理可以从剩余组件中重建已删除的秘密信息，并且在相当大比例的案例中实现了精确重建。研究强调，删除应在代理和内存系统层面进行行为评估，而不仅仅是作为存储操作，并探讨了诸如溯源感知闭包等方法来缓解此问题。 AI

影响凸显了 LLM 代理中潜在的隐私风险，需要对数据删除进行行为审计。

排序理由学术论文，介绍了一个新基准和研究结果。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 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: 4 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 学术论文，介绍了一个新基准和研究结果。[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: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CL TIER_1 English(EN) · Ehsan Saleh, Roozbeh Nahavandi, Sama Salarian, Raef Bassily, Andrew Perrault · 2026-10-23 04:00

AfterDeleteBench: Auditing Post-Deletion Information Persistence in LLM Agents

arXiv:2610.31366v1 Announce Type: cross Abstract: Many LLM agents maintain external persistent memory and derive additional state from past interactions, such as summaries. Deleting an original memory record therefore need not remove information carried by its descendants. We int…

本文如何被排名

报道来源 [1]

AfterDeleteBench: Auditing Post-Deletion Information Persistence in LLM Agents

相关实体

相关话题