MemORAI: Memory Organization and Retrieval via Adaptive Graph Intelligence for LLM Conversational Agents
Hung Pham Van, Nguyen Manh Hieu, Khang Pham Tran Tuan, Nam Le Hai, Linh Ngo Van, Diep Thi-Ngoc Nguyen, Trung Le · 2026
Read MemORAI: Memory Organization and Retrieval via Adaptive Graph Intelligence for LLM Conversational Agents by Hung Pham Van, Nguyen Manh Hieu, Khang Pham Tran Tuan, Nam Le Hai, Linh Ngo Van, Diep Thi-Ngoc Nguyen, Trung Le (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-01386.txt.
Description
Hung Pham Van and six co-authors build a memory system that lets a language model agent recall users across long chats. Plain models forget, and earlier graph-based memories thin out facts, lose track of where facts came from, and search the same way for every query.
MemORAI adds three parts. A filter with two compression layers keeps only facts tied to the user's profile. A rich graph stores relations between facts and tracks each fact's origin down to the chat turn. Retrieval adapts to the query, weighting graph links by what is asked and ranking with a tuned PageRank method. On the LOCOMO and LongMemEval benchmarks, the system tops prior work at both recall and tailored replies, which the authors read as proof that selective storage and adaptive search both matter.
Catalog record
| CREATOR | Hung Pham Van, Nguyen Manh Hieu, Khang Pham Tran Tuan, Nam Le Hai, Linh Ngo Van, Diep Thi-Ngoc Nguyen, Trung Le |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 52 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.01386