arXiv CS AI
techLean Left
EdgeMem: LLM-Free Agent Memory Construction and Retrieval via Evidence-Preserving Multi-Anchor Hypergraphtranslating…
arXiv:2609.05553v1 Announce Type: new
Abstract: Agent memory allows LLM agents to use earlier interactions when answering new queries. Existing methods often compress interaction histories into summaries or other LLM-generated representations. Repeated generation adds cost and…
Keywords#Agent Memory#Agent Memory Construction#EdgeMem#Evidence-Preserving#Evidence-Preserving Multi-Anchor