From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space
Quick Answer
NapMem introduces a structured action space for long-term user memory in conversational agents, enhancing memory navigation over passive retrieval.
Quick Take
Experiments demonstrate its competitive performance across memory-intensive tasks while maintaining general reasoning abilities, suggesting a significant advancement in personalized AI interactions.
Key Points
- NapMem organizes user memory into a multi-granularity memory pyramid.
- The agent selects memory based on queries and intermediate evidence.
- Competitive performance observed on PersonaMem-v2 and LongMemEval benchmarks.
- Maintains general reasoning abilities in non-memory tasks.
- Analysis covers storage, inference costs, and behavior.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Long-term user memory is essential for personalized conversational agents, yet many memory systems still expose memory through passive retrieval interfaces, making the model a consumer of pre-selected evidence. We introduce NapMem, a framework for learning to use long-term user memory as a structured action space rather than passively retrieved context. NapMem organizes user history into a linked multi-granularity memory pyramid, where raw conversations, typed memory records, topic tracks, and user profiles are connected through provenance relations, and exposes these levels through memory tools. The agent is trained to select memory according to the query and intermediate evidence, allowing it to inspect different memory granularities before answering. Experiments on PersonaMem-v2, LongMemEval, and LoCoMo show that a NapMem agent trained with memory-tool reinforcement learning is competitive across diverse memory-intensive tasks, while evaluations on non-memory tasks suggest that the learned policy largely preserves general reasoning and tool-use abilities. Additional analyses examine storage, inference cost, tool-use behavior, and ablations over navigation, memory granularity, and RL training. Our results suggest that long-term user memory benefits from coupling structured storage with a learned policy for using memory at the appropriate granularity.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2607.05794 [cs.AI] |
| (or arXiv:2607.05794v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2607.05794 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yue Xu [view email]
[v1]
Tue, 7 Jul 2026 03:47:12 UTC (1,199 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →HOBA: Hierarchical On-Policy Bidding Agents for Adaptive Online Advertising
HOBA (Hierarchical On-policy Bidding Agents) is a novel hierarchical reinforcement learning framework that enhances online advertising bidding systems by improving adaptability and reducing hyperparameter tuning costs. It utilizes a for hyperparameter inference, a SARSA agent for expert model selection, and a dynamic expert pool for bid execution, achieving a +3.6% increase in target cost during large-scale deployment and outperforming state-of-the-art baselines on AuctionNet.