持久化之前先限定范围：防止智能体记忆中的跨族干扰
Scope Before You Persist: Preventing Cross-Family Interference in Agent Memory
- Northwestern University（西北大学）
- Pinterest, Inc.（Pinterest 公司）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本研究提出范围匹配方法，通过将检索范围与认证范围对齐，防止持久记忆中的跨族干扰，在ProcStream-RSI上显著提升智能体轨迹效用并消除有害更新。
AI中文摘要：
持久化记忆使语言模型智能体无需更新模型权重即可改进提示和技能。我们证明，将检索范围与认证范围相匹配，能使这些编辑在重复出现的任务族中支持可靠的重复适应。我们在ProcStream-RSI（一个12轮代码修复流）上研究冻结模型智能体，使用正交回归控制（ORC）——一种针对持久技能编辑的基于执行的门控。在一项保持提案和门控决策固定的干预中，仅针对其来源族检索每个已接受的技能，将平均隐藏轨迹效用从全局记忆下的0.713提高到0.816，并将有害部署从八个中的六个减少到零。在27个配对随机顺序流中，Scoped-ORC相对于Global-ORC将平均轨迹效用提高了0.063 [0.037, 0.094]，接受了63次而非12次更新，并在19/27个流中产生了多个已接受的更新，且0/63次接受是有害的。全局控制达到0.713，低于静态智能体的0.775，因为局部有效的编辑可能干扰不相关的族。这些结果确立了范围匹配作为持久智能体记忆的补充控制：认证决定编辑是否受支持，而检索范围决定该证据在何处授权其使用。
英文摘要：
Persistent memory lets language-model agents improve prompts and skills without updating model weights. We show that matching retrieval scope to certification scope enables these edits to support reliable repeated adaptation across recurring task families. We study frozen-model agents on ProcStream-RSI, a 12-round code-repair stream, using Orthogonal Regression Control (ORC), an execution-grounded gate for persistent skill edits. In an intervention that holds proposals and gate decisions fixed, retrieving each accepted skill only for its originating family raises mean hidden trajectory utility from 0.713 under global memory to 0.816 and changes harmful deployments from six of eight to none. In 27 paired randomized-order streams, Scoped-ORC improves mean trajectory utility by 0.063 [0.037, 0.094] over Global-ORC, accepts 63 rather than 12 updates, and produces multiple accepted updates in 19/27 streams, with 0/63 harmful acceptances. The global control reaches 0.713, below the static agent's 0.775, because locally valid edits can interfere with unrelated families. These results establish scope matching as a complementary control for persistent agent memory: certification determines whether an edit is supported, while retrieval scope determines where that evidence authorizes its use.
