AI RESEARCH
Personalize-then-Store: Benchmarking and Learning Personalized Memory for Long-horizon Agents
arXiv CS.AI
•
ArXi:2605.25535v1 Announce Type: new Existing large language model (LLM) based memory systems apply universal, static policies that overlook a fundamental reality: the contexts that are worth storing in memory are different across users. This misalignment wastes limited memory budget on transient interactions while failing to preserve critical context for long horizon tasks. To address this gap, we investigate an underexplored question: can LLM based memory systems learn personalized memory policies? We.