SwiftMem: Fast Agentic Memory via Query-aware Indexing
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908761400016896 |
|---|---|
| author | Tian, Anxin Li, Yiming Li, Xing Zhen, Hui-Ling Chen, Lei Yu, Xianzhi Dong, Zhenhua Yuan, Mingxuan |
| author_facet | Tian, Anxin Li, Yiming Li, Xing Zhen, Hui-Ling Chen, Lei Yu, Xianzhi Dong, Zhenhua Yuan, Mingxuan |
| contents | Agentic memory systems have become critical for enabling LLM agents to maintain long-term context and retrieve relevant information efficiently. However, existing memory frameworks suffer from a fundamental limitation: they perform exhaustive retrieval across the entire storage layer regardless of query characteristics. This brute-force approach creates severe latency bottlenecks as memory grows, hindering real-time agent interactions. We propose SwiftMem, a query-aware agentic memory system that achieves sub-linear retrieval through specialized indexing over temporal and semantic dimensions. Our temporal index enables logarithmic-time range queries for time-sensitive retrieval, while the semantic DAG-Tag index maps queries to relevant topics through hierarchical tag structures. To address memory fragmentation during growth, we introduce an embedding-tag co-consolidation mechanism that reorganizes storage based on semantic clusters to improve cache locality. Experiments on LoCoMo and LongMemEval benchmarks demonstrate that SwiftMem achieves 47$\times$ faster search compared to state-of-the-art baselines while maintaining competitive accuracy, enabling practical deployment of memory-augmented LLM agents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_08160 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SwiftMem: Fast Agentic Memory via Query-aware Indexing Tian, Anxin Li, Yiming Li, Xing Zhen, Hui-Ling Chen, Lei Yu, Xianzhi Dong, Zhenhua Yuan, Mingxuan Computation and Language Artificial Intelligence Agentic memory systems have become critical for enabling LLM agents to maintain long-term context and retrieve relevant information efficiently. However, existing memory frameworks suffer from a fundamental limitation: they perform exhaustive retrieval across the entire storage layer regardless of query characteristics. This brute-force approach creates severe latency bottlenecks as memory grows, hindering real-time agent interactions. We propose SwiftMem, a query-aware agentic memory system that achieves sub-linear retrieval through specialized indexing over temporal and semantic dimensions. Our temporal index enables logarithmic-time range queries for time-sensitive retrieval, while the semantic DAG-Tag index maps queries to relevant topics through hierarchical tag structures. To address memory fragmentation during growth, we introduce an embedding-tag co-consolidation mechanism that reorganizes storage based on semantic clusters to improve cache locality. Experiments on LoCoMo and LongMemEval benchmarks demonstrate that SwiftMem achieves 47$\times$ faster search compared to state-of-the-art baselines while maintaining competitive accuracy, enabling practical deployment of memory-augmented LLM agents. |
| title | SwiftMem: Fast Agentic Memory via Query-aware Indexing |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2601.08160 |