MemoType：按记忆类型分别检索，Recall@1 最高提升 16.18%
做智能体记忆的可以看看：别再用一套 embedding 搜所有记忆了，先分类再按类型检索，三个数据集上 Recall@1 最多涨 16.18%，还附了 TriMEM 数据集。
DAIR.AI 分享了一篇关于智能体记忆的论文。论文指出，当记忆库包含多种类型时，任何单一检索策略的期望精度都有上限。MemoType 会先对每条记忆和每个查询做类型分类，再按类型匹配并应用对应的检索策略。作者同时发布了带类型标签的数据集 TriMEM，用于训练分类器。在三个数据集上，Recall@1 最高提升 16.18%。
Nice paper on agent memory.
It explores an effective routing mechanism for memory in agents.
The overall finding is that you want a separate retrieval strategy for each type of memory instead of one embedding search over everything.
The authors prove that any single retrieval strategy has an upper bound on expected precision when the memory store holds several types.
MemoType classifies each memory and each query by type, retrieves memories that match the query type, and applies the strategy for that type. A new dataset, TriMEM, provides the type labels needed to train the classifier.
Recall@1 improves by up to 16.18% across three datasets.
Paper: https://t.co/WFiGbPsaLe
