From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms

TL;DR

This survey introduces an evolutionary framework for LLM agent memory mechanisms, formalizing their development into three stages and analyzing key drivers and transformative mechanisms to guide future design.

Contribution

It provides a unified theoretical framework and design principles for the evolution of LLM agent memory systems, bridging engineering and cognitive science perspectives.

Findings

Formal definition of Storage, Reflection, and Experience stages

Analysis of three core drivers: long-range consistency, dynamic challenges, continual learning

Identification of proactive exploration and cross-trajectory abstraction as key mechanisms

Abstract

Large Language Model (LLM)-based agents have fundamentally reshaped artificial intelligence by integrating external tools and planning capabilities. While memory mechanisms have emerged as the architectural cornerstone of these systems, current research remains fragmented, oscillating between operating system engineering and cognitive science. This theoretical divide prevents a unified view of technological synthesis and a coherent evolutionary perspective. To bridge this gap, this survey proposes a novel evolutionary framework for LLM agent memory mechanisms, formalizing the development process into three stages: Storage (trajectory preservation), Reflection (trajectory refinement), and Experience (trajectory abstraction). We first formally define these three stages before analyzing the three core drivers of this evolution: the necessity for long-range consistency, the challenges in dynamic environments, and the ultimate goal of continual learning. Furthermore, we specifically explore two transformative mechanisms in the frontier Experience stage: proactive exploration and cross-trajectory abstraction. By synthesizing these disparate views, this work offers robust design principles and a clear roadmap for the development of next-generation LLM agents.