MoM：记忆的记忆

中文摘要 AI 辅助

对于长时程LLM智能体而言，记忆问题不在于曾经记录了些什么，而在于什么内容\u201c当前仍然有效\u201d。大多数设计只是间接地回答这个问题：每一次交互都被存储，当前状态在查询时通过检索和协调记录来重建，因此过时的值会重新进入，相同的冲突会被反复重新裁定。在写入时提交当前值可以避免这种情况，但现有的写入时（CRUD）记忆会覆盖，因此错误的更新无法恢复，先前的状态会丢失。我们采取了缺失的组合——\u201c到达即提交，同时保留被替换的内容\u201d——并将其形式化为\u201c记忆的记忆\u201d（MoM）：记忆不仅跟踪内容，还跟踪其自身条目的来源、状态和历史。我们将MoM实例化为\u201c来源记忆\u201d（P-Mem），一个类型化的来源图，其\u201c活跃前沿\u201d为每个已解析的键暴露一个当前值，而被替换的值则作为来源保留；类型化操作决定新的观察是支持、取代、竞争、拒绝、撤销还是解析现有值。P-Mem的决定性优势在于有效性而非准确性：其回合级读取在准确率上与最强的检索记忆相当，而读取令牌数减少了约4倍——这是一种检索粒度效应——同时图引导的回合剪枝将知识更新的过时回答率从19.4%降至10.9%；在修订链上，它保持100%的准确率，而查询时读取则降至25%；并且，由于被替换的值被保留而非覆盖，它能恢复CRUD记忆无法恢复的已提交错误（100%对0%）。

英文摘要

For a long-horizon LLM agent, the memory question is not what was once recorded but what \emph{currently holds}. Most designs answer it only indirectly: every interaction is stored, and the present is reconstructed at query time by retrieving and reconciling records, so stale values re-enter and the same conflicts are re-litigated. Committing the current value at write time avoids this, but existing write-time (CRUD) memories overwrite, so a wrong update is unrecoverable and prior state is lost. We take the missing combination---\emph{commit on arrival while retaining what is displaced}---and formalize it as \textsc{Memory of Memory} (MoM): memory tracks not only content but the provenance, status, and history of its own entries. We instantiate MoM as \textsc{Provenant Memory} (P-Mem), a typed provenance graph whose \emph{active frontier} exposes one current value per resolved key while displaced values are retained as provenance; typed operations decide whether a new observation supports, supersedes, contests, rejects, revokes, or resolves an existing value. P-Mem's decisive gain is validity rather than accuracy: its turn-level read matches the strongest retrieval memory in accuracy at $\sim$4$\times$ fewer read tokens---a retrieval-granularity effect---while graph-guided turn pruning cuts the knowledge-update stale-answer rate (19.4\%$\rightarrow$10.9\%); on revision chains it stays at 100\% where query-time reading collapses to 25\%, and, because displaced values are retained rather than overwritten, it recovers committed errors a CRUD memory cannot (100\% vs.\ 0\%).

发表机构

• National University of Singapore（新加坡国立大学）

机构由 AI 辅助整理，请以论文原文为准。