AI RESEARCH
TiMem: Temporal-Hierarchical Memory Consolidation for Long-Horizon Conversational Agents
arXiv CS.AI
•
ArXi:2601.02845v2 Announce Type: replace-cross Long-horizon conversational agents have to manage ever-growing interaction histories that quickly exceed the finite context windows of large language models (LLMs). Existing memory frameworks provide limited for temporally structured information across hierarchical levels, often leading to fragmented memories and unstable long-horizon personalization.