用二进制签名和压缩矩阵实现超快记忆检索，适合长期智能体。
Hippocampus: An Efficient and Scalable Memory Module for Agentic AI
- 用二进制签名和无损令牌流构建记忆，避免向量计算
- 检索延迟降低31倍，每查询令牌量减少14倍
- 线性可扩展，适合长时智能体部署
智能体需要持久记忆来存储用户历史，但现有系统依赖密集向量库或知识图谱遍历，存在检索延迟高、存储难扩展的问题。我们提出Hippocampus，采用紧凑的二进制签名进行语义搜索，结合无损的令牌ID流实现内容精确重建。其核心是动态小波矩阵（DWM），将两种数据流压缩并联合索引，支持在压缩域内超快搜索，无需昂贵的向量或图计算。该设计随内存线性扩展，适用于长周期智能体应用。实验表明，Hippocampus在LoCoMo和LongMemEval基准上保持准确率的同时，端到端检索延迟降低最多31倍，每查询令牌开销减少最多14倍。
原文摘要 · Abstract (English)
Agentic AI require persistent memory to store user-specific histories beyond the limited context window of LLMs. Existing memory systems use dense vector databases or knowledge-graph traversal (or hybrid), incurring high retrieval latency and poor storage scalability. We introduce Hippocampus, an agentic memory management system that uses compact binary signatures for semantic search and lossless token-ID streams for exact content reconstruction. Its core is a Dynamic Wavelet Matrix (DWM) that compresses and co-indexes both streams to support ultra-fast search in the compressed domain, thus avoiding costly dense-vector or graph computations. This design scales linearly with memory size, making it suitable for long-horizon agentic deployments. Empirically, our evaluation shows that Hippocampus reduces end-to-end retrieval latency by up to 31$\times$ and cuts per-query token footprint by up to 14$\times$, while maintaining accuracy on both LoCoMo and LongMemEval benchmarks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。