跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.20927 · v1

LEDGER: Auditable Agent Memory Under Change

首次提交 2026-11-16 19:37
版本更新 2026-11-16 19:37
雷达观测 2026-11-17 17:11
arXiv ↗PDF ↗

摘要

Agent memory systems are mostly evaluated on fixed histories, and only their final answers are scored. In real use, facts change, sources differ in reliability, and users ask the agent to forget things. We present LEDGER, a memory layer in which a language model proposes facts and explicit rules decide which value is current, which source to trust, and what to delete. Each fact keeps its source and the period in which it held. LEDGER+EP adds a search over raw conversation turns for questions the ledger does not cover, and scrubs these turns on deletion. We compare both against no memory, full-history replay, retrieval over raw transcripts, and Mem0, with the same answering model and prompt, on a synthetic multi-session coding-agent workload and a 210-question subset of LongMemEval, and we score both the evidence each memory sends to the model and the answer. On the coding workload, LEDGER+EP in fallback mode answers 94.3% of questions correctly on five development seeds and 91.1% on five held-out seeds, against 82.1% and 83.3% for full replay, the strongest baseline. When a deletion request only describes the fact, no deleted value reached our systems' evidence in any evaluated probe. On LongMemEval, LEDGER+EP in augment mode has the same observed accuracy as full replay (74.8%) with 1% of its context. The main limitation is that the coding workload is synthetic and built by us; on LongMemEval, the differences between the strongest systems are within the confidence intervals.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-16 19:37 · 最新观测版本

LEDGER: Auditable Agent Memory Under Change

Agent memory systems are mostly evaluated on fixed histories, and only their final answers are scored. In real use, facts change, sources differ in reliability, and users ask the agent to forget things. We present LEDGER, a memory layer in which a language model proposes facts and explicit rules decide which value is current, which source to trust, and what to delete. Each fact keeps its source and the period in which it held. LEDGER+EP adds a search over raw conversation turns for questions the ledger does not cover, and scrubs these turns on deletion. We compare both against no memory, full-history replay, retrieval over raw transcripts, and Mem0, with the same answering model and prompt, on a synthetic multi-session coding-agent workload and a 210-question subset of LongMemEval, and we score both the evidence each memory sends to the model and the answer. On the coding workload, LEDGER+EP in fallback mode answers 94.3% of questions correctly on five development seeds and 91.1% on five held-out seeds, against 82.1% and 83.3% for full replay, the strongest baseline. When a deletion request only describes the fact, no deleted value reached our systems' evidence in any evaluated probe. On LongMemEval, LEDGER+EP in augment mode has the same observed accuracy as full replay (74.8%) with 1% of its context. The main limitation is that the coding workload is synthetic and built by us; on LongMemEval, the differences between the strongest systems are within the confidence intervals.

查看此版本解读 →查看此版本原文 ↗