---
id: 20261030-T0-20
title: "当检索不够时：Agent记忆中的过时事实"
title_en: "When retrieval falls short: stale facts in agent memory"
url: https://ai.daily.yangsir.net/daily/20261030-T0-20
issue_date: 2026-10-30
publish_date: 2026-10-29T04:00:00.000Z
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2610.36308
---

# 当检索不够时：Agent记忆中的过时事实

arXiv:2610.36308v1 公告类型：新论文。摘要：记忆型Agent的基准测试通常将检索与答案生成合并为一个分数，导致难以判断失败究竟源于遗漏了更新，还是源于未使用已检索到的更新。我们在sing上考察了这一区分。

## English Version

**When retrieval falls short: stale facts in agent memory**

arXiv:2610.36308v1, announcement type: new paper. Abstract: Benchmarks for memory-based agents typically merge retrieval and answer generation into a single score, making it hard to tell whether a failure stems from missing an update or from failing to use a retrieved update. We examine this distinction on sing.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2610.36308)

**详情页**：https://ai.daily.yangsir.net/daily/20261030-T0-20

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*