Meta FAIR 提出 Routed Continual Learning 改进智能体记忆
Meta FAIR 的新论文，智能体任务混着跑时全局记忆基本没用，改成按任务簇路由记忆后奖励高 15%-30%，成本还降了 65% 以上。
Meta FAIR 的论文指出，用单一全局历史让智能体学习过往经验，在任务交错场景下相比没有记忆平均只有 +0.015 的归一化奖励提升。论文提出 Routed Continual Learning：将历史轨迹聚类，每个簇由固定库中的一个记忆系统保存状态，每次运行前由 LLM 路由器决定复用旧簇或新建簇。在混合 301 个 CL-Bench 实例与 97 个 τ³-Bench 银行实例的流上测试四个语言模型，相比全局历史奖励提升最高可得奖励的 15% 至 30%，API 成本下降 65% 至 72%。实验为顺序运行，并行智能体更新同一簇的场景尚未验证。
Brilliant paper on agent memory from FAIR at Meta.
(bookmark it)
A single global history is the usual way to let agents learn from past runs. When tasks are interleaved, it gains almost nothing over no memory at all, an average of +0.015 in normalized reward.
This work proposes Routed Continual Learning. Past trajectories are grouped into clusters, and each cluster keeps its state in one memory system from a fixed library. Before each run, an LLM router reuses a cluster or creates a new one.
They tested it on a stream mixing 301 CL-Bench instances with 97 banking instances from τ³-Bench, across four language models.
Routed CL raised reward over the global history by 15% to 30% of the maximum attainable reward. Estimated API cost fell by 65% to 72%.
The runs were sequential, so parallel agents updating one cluster remain untested.
If your agents handle mixed work, it might help to split their memory into clusters before adding a bigger memory system.
https://t.co/gs8VC9jYCN
Chat with Paper: https://t.co/iufIkWUgcO
