MemORAI: Memory Organization and Retrieval via Adaptive Graph Intelligence for LLM Conversational Agents
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915974225068032 |
|---|---|
| author | Van, Hung Pham Hieu, Nguyen Manh Tuan, Khang Pham Tran Hai, Nam Le Van, Linh Ngo Diep, Nguyen Thi Ngoc Le, Trung |
| author_facet | Van, Hung Pham Hieu, Nguyen Manh Tuan, Khang Pham Tran Hai, Nam Le Van, Linh Ngo Diep, Nguyen Thi Ngoc Le, Trung |
| contents | Large Language Models (LLMs) lack persistent memory for long-term personalized conversations. Existing graph-based memory systems suffer from information dilution, absent provenance tracking, and uniform retrieval that ignores query context. We introduce MemORAI (Memory Organization and Retrieval via Adaptive Graph Intelligence), a framework that integrates three innovations: selective memory filtering with dual-layer compression to retain user-persona-relevant content, a provenance-enriched multi-relational graph tracking factual origins at the turn level, and query-adaptive subgraph retrieval with Dynamic Weighted PageRank that applies query-conditioned edge weighting. Evaluated on LOCOMO and LongMemEval benchmarks, MemORAI achieves state-of-the-art performance in memory retrieval and personalized response generation, demonstrating that selective storage, enriched representation, and adaptive retrieval are essential for coherent, personalized LLM agents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_01386 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MemORAI: Memory Organization and Retrieval via Adaptive Graph Intelligence for LLM Conversational Agents Van, Hung Pham Hieu, Nguyen Manh Tuan, Khang Pham Tran Hai, Nam Le Van, Linh Ngo Diep, Nguyen Thi Ngoc Le, Trung Computation and Language Large Language Models (LLMs) lack persistent memory for long-term personalized conversations. Existing graph-based memory systems suffer from information dilution, absent provenance tracking, and uniform retrieval that ignores query context. We introduce MemORAI (Memory Organization and Retrieval via Adaptive Graph Intelligence), a framework that integrates three innovations: selective memory filtering with dual-layer compression to retain user-persona-relevant content, a provenance-enriched multi-relational graph tracking factual origins at the turn level, and query-adaptive subgraph retrieval with Dynamic Weighted PageRank that applies query-conditioned edge weighting. Evaluated on LOCOMO and LongMemEval benchmarks, MemORAI achieves state-of-the-art performance in memory retrieval and personalized response generation, demonstrating that selective storage, enriched representation, and adaptive retrieval are essential for coherent, personalized LLM agents. |
| title | MemORAI: Memory Organization and Retrieval via Adaptive Graph Intelligence for LLM Conversational Agents |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2605.01386 |