Hippocampus: An Efficient and Scalable Memory Module for Agentic AI

TL;DR

Hippocampus is a scalable, efficient memory module for agentic AI that uses compact signatures and a Dynamic Wavelet Matrix to enable fast, low-latency retrieval and scalable storage for long-term memory tasks.

Contribution

The paper introduces Hippocampus, a novel memory system that significantly improves retrieval speed and storage efficiency using compact signatures and a Dynamic Wavelet Matrix.

Findings

Reduces retrieval latency by up to 31 times.

Cuts per-query token footprint by up to 14 times.

Maintains accuracy on key benchmarks.

Abstract

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.