Fetching the paper…

MemBench: Towards More Comprehensive Evaluation on the Memory of LLM-based Agents · Around