Remember Me, Refine Me: A Dynamic Procedural Memory Framework for Experience-Driven Agent Evolution

This paper proposes ReMe, a dynamic procedural memory framework for experience-driven agent evolution. It addresses the limitations of static memory in LLM agents by introducing multi-faceted distillation, context-adaptive reuse, and utility-based refinement mechanisms. ReMe enables agents to internalize "how-to" knowledge, reduce redundant trial-and-error, and achieve continuous improvement without expensive parameter retraining, leading to more efficient and adaptable AI systems.