1 paper
Runzhe Wang, Huilin Lu, Shengjie Liu +2
Large Language Models (LLMs) show promise as tool-using agents but remain limited in long-horizon tasks that require remembering, organizing, and reusing knowledge. Prior memory ap…