MemLens: A Value-Aware Memory Management System with Interactive Analytics for LLM-based Agents
MemLens ranks agent memories by value, then lets users inspect how those records affect quality, latency, and token use.
The paper says current LLM memory systems often keep interaction records without judging their utility, allowing redundant or low-impact data to persist. MemLens treats memory records as first-class data objects and exposes their lifecycle through an interactive analytics dashboard. Its workflow includes Shapley-style memory evaluation, value-aware storage, and memory-assisted response. The authors demonstrate it in a study-copilot application for comparing memory strategies.
ArXiv · AI/CL/LG's note
The paper says current LLM memory systems often keep interaction records without judging their utility, allowing redundant or low-impact data to persist. MemLens treats memory records as first-class data objects and exposes their lifecycle through an interactive analytics dashboard. Its workflow includes Shapley-style memory evaluation, value-aware storage, and memory-assisted response. The authors demonstrate it in a study-copilot application for comparing memory strategies.
ArXiv · AI/CL/LG's note
score 4