Hippocampus: An Efficient and Scalable Memory Module for Agentic AI
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910022142787584 |
|---|---|
| author | Li, Yi Cao, Lianjie Ahmed, Faraz Sharma, Puneet Li, Bingzhe |
| author_facet | Li, Yi Cao, Lianjie Ahmed, Faraz Sharma, Puneet Li, Bingzhe |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_13594 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Hippocampus: An Efficient and Scalable Memory Module for Agentic AI Li, Yi Cao, Lianjie Ahmed, Faraz Sharma, Puneet Li, Bingzhe Artificial Intelligence 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. |
| title | Hippocampus: An Efficient and Scalable Memory Module for Agentic AI |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2602.13594 |