Meta 论文：智能体记忆按问题类型分池比共用全量历史更有效
Meta 论文：智能体记忆按问题类型分池更有效
Meta 新论文提出 Routed CL，用 LLM 路由器在每次运行前复用或新建记忆簇，每簇采用 5 种记忆类型之一（如原始历史运行或蒸馏经验）。在 398 个混合问题和 4 个模型上，共用历史仅比无记忆平均高 0.015（满分 1），Routed CL 则达到最高可能分数的 15% 至 30%，并将预估 API 成本降低 65% 至 72%。
Rohan Paul· @rohanpaul_ai · X·2026-11-03 02:05· 1 小时前AI 评分44
AI 导读
Meta 新论文发现，AI 智能体从过往尝试中学习时，把新问题匹配到相似问题的笔记，比共用一份全量历史记录效果更好。其提出的 Routed CL 用一个 LLM 路由器在每次运行前复用或新建记忆簇，每簇采用 5 种记忆类型之一（如原始历史运行或蒸馏经验）。
正文 · 原文
New Meta paper finds that AI agents improve more from their past attempts when each new problem is matched to notes from similar problems, instead of one shared record of everything they have done.
A simple way to let agents learn from past runs is one shared history. With mixed tasks, it gets costly and mixes unrelated experience.
Routed CL adds an LLM router that reuses a memory cluster or starts a new one before each run. Each cluster uses 1 of 5 memory types, such as raw past runs or distilled lessons.
Across 398 mixed problems and 4 models, the shared history barely beat no memory, with a mean gain of 0.015 out of 1. Routed CL beat it by 15% to 30% of the maximum possible score and cut estimated API cost by 65% to 72%.
If your agents handle mixed task types, split their memory into separate pools by problem type instead of sharing one history.
来源：Rohan Paul · x.com
来源：RadarAI · aihot.news