跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+1208新增论文
+950其中 AI 相关
+153已有论文修订
+59新增 AI 导读

arXiv 最近完成扫描:2026-11-18 04:28 · 全站最近入库:2026-11-18 14:10各信源进度 →

统计口径与数据新鲜度

统计 2026-11-17 14:35 — 2026-11-18 14:35(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-18 04:28最近发现新增版本:2026-11-18 14:10库内论文最新更新时间:2026-11-18 08:00

每分钟刷新已保存的数据。空检查不代表采集异常;新收录可能包含历史论文。AI 导读按新增版本计数。

26,057 已收录论文21,704 AI 相关7,828 有修订112 HF 社区发现

浏览论文库

时间范围跟随排序:版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常,查看状态 →

找到 1 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-18 04:27(北京时间)
版本更新 · v3arXiv:2609.22818

The Price of Safety: Benign-Case Utility and Token Overhead of Memory-Poisoning Defenses in LLM Agents

Pritom Bhowmik

原文摘要

Memory-poisoning defenses for LLM agents are usually judged by how well they stop attacks, yet almost all the traffic an agent sees is benign. We measure what these defenses cost when no attack is present. Our harness holds the backbone model, memory backend, retrieval, answer prompt, and judge fixed and changes only the defense. Both LLM-based defenses run on the agent's own backbone, and an integrity check verifies every run. We test four defenses on half of LoCoMo: five long conversations (750 turns and 255 questions per run), with each condition run three times. Sanitization and provenance checking never fired. LLM-based anomaly detection blocked 9 of 2,250 turns (0.4%). A read-time reranker dropped 2.2% of the memories it judged, withholding at least one memory from 14.9% of questions. A blind audit found content-based justification for only 2 of 50 sampled drops. No defense showed a detectable accuracy cost, at 95% intervals of $\pm$2.4 to $\pm$4.6 points; the reranker changed accuracy by +0.6 points (95% CI [-2.6, +3.8]). Anomaly detection and the stacked condition gained 4.7 points, but a blocked turn changes the memories extracted after it, so we do not count this as a benefit. Token overhead was 0% for sanitization and provenance, 1.5% for anomaly detection, 3.0% for the reranker, and 4.5% with all four stacked. An earlier version of this study ran the reranker condition on a different model than its baseline. A baseline on that same model shows the swap produced 2.9 of the 4.4 points of accuracy cost we then reported. Attack success rate reveals none of this, so we report benign cost alongside it. We release the harness, labels, and per-run outputs.

cs.CR · cs.AI