Meta论文：路由记忆机制提升AI Agent学习效率
New Meta paper finds that AI agents improve more from their past attempts when e…
Agent记忆管理是落地关键痛点，这篇论文给出了具体的路由架构与量化收益，对构建高效多任务Agent有直接参考价值。
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.
Meta 最新论文发现，当每个新问题都与类似问题的笔记相匹配时，AI 智能体能从过去的尝试中获得更多提升，而不是依赖一个记录所有行为的共享历史。
A simple way to let agents learn from past runs is one shared history. With mixed tasks, it gets costly and mixes unrelated experience.
让智能体从过往运行中学习的简单方法是使用一个共享历史记录。但在混合任务中，这种方法成本高昂且会混杂不相关的经验。
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.
Routed CL 增加了一个 LLM 路由器，在每次运行前重用现有的记忆集群或创建一个新的。每个集群使用 5 种记忆类型之一，例如原始过往运行记录或提炼后的经验教训。
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%.
在 398 个混合问题和 4 个模型上测试，共享历史记录的表现仅略优于无记忆状态，平均增益为 1 分中的 0.015 分。而 Routed CL 比其高出最大可能得分的 15% 至 30%，并将预估 API 成本降低了 65% 至 72%。
If your agents handle mixed task types, split their memory into separate pools by problem type instead of sharing one history.
如果你的智能体处理多种任务类型，应按问题类型将它们的记忆拆分为独立的池，而不是共享单一的历史记录。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力