A survey on the memory mechanism of large language model based agents. May 4, 2025 · Abstract.
- A survey on the memory mechanism of large language model based agents. Mar 27, 2025 · Large Language Model (LLM) agents, with goal-driven behaviors and dynamic adaptation capabilities, potentially represent a critical pathway toward artificial general intelligence. Abstract Large language model (LLM) based agents have recently attracted much attention from the research and industry communities. Then, we systematically review previous studies on how to design and evaluate the memory module. Jan 21, 2024 · Large Language Models (LLMs) have achieved remarkable success across a wide array of tasks. Compared with original LLMs, LLM- based agents are featured in their self-evolving capability, which is the basis for solving real-world problems that need long-term and complex agent-environment interactions. Feb 10, 2025 · dblp: A Survey on the Memory Mechanism of Large Language Model based Agents. 2 Surveys on Large Language Model-based Agents Based on the capability of LLMs, people have conducted a lot of studies on building LLM-based agents, which can autonomously perceive environments, take actions, accumulate knowledge, and evolve themselves. We categorize the core components of these agents into three modules: the perception module for multimodal input integration, the memory module for storing and retrieving evolving knowledge, and the action Jul 2, 2025 · To bridge this gap, in this paper, we propose a comprehensive survey on the memory mechanism of LLM-based agents. Apr 21, 2024 · The paper presents a comprehensive survey of memory mechanisms in LLM-based agents, detailing the writing, managing, and reading phases. Apr 22, 2024 · This paper provides a comprehensive survey of the memory mechanisms used in large language model-based agents. Apr 23, 2025 · Second, we systematically organize existing memory-related work and propose a categorization method based on three dimensions (object, form, and time) and eight quadrants. pdf https: //github. To bridge this gap, in this paper, we propose a comprehensive survey on the memory mechanism of LLM-based agents. com/nuster1128/LLM_Agent_Memory_Survey 推荐阅读 • 对齐LLM偏好的直接偏好优化方法：DPO、IPO、KTO • 2024：ToB、Agent、多模态 • TA们的RAG真正投产了吗？ （上） Mar 27, 2025 · The era of intelligent agents is upon us, driven by revolutionary advancements in large language models. Large language model (LLM) based agents have recently attracted much attention from the research and industry communities. The authors propose a unified framework for designing LLM-based agents, encompassing profiling, memory, planning, and action modules. This survey systematically deconstructs LLM agent systems through a methodology-centered taxonomy, linking Nov 18, 2024 · 一篇survey总结LLM记忆的，实现上总结了三个要素：Source，Forms，Operations。 _a survey on the memory mechanism of large language model based agents この論文「A Survey on the Memory Mechanism of Large Language Model based Agents」は、最近注目を集めている大規模言語モデル（LLM）ベースのエージェントにおける記憶メカニズムに関する包括的な調査を提供しています。本論文は、LLMベースのエージェントが長期的かつ複雑なエージェントと環境の相互作用に Large language model (LLM) based agents have recently attracted much attention from the research and industry communities. Despite the proposal of many advanced memory models in recent research, however, there remains a lack of unified implementations under a general framework . org/pdf/2404. Compared with original LLMs, LLM-based agents are featured in their self-evolving capability, which is the basis for solving real-world problems that need long-term and complex agent A Survey on the Memory Mechanism of Large Language Model based Agents https: //arxiv. The key component to support agent-environment interactions is The paper list of the 86-page paper "The Rise and Potential of Large Language Model Based Agents: A Survey" by Zhiheng Xi et al. [ 2404. A Survey on the Memory Mechanism of Large Language Model based Agents (2024) This paper comprehensively surveys LLM-based agents' memory mechanisms, reviewing design and evaluation, presenting applications, and suggesting future directions. As a critical part, their memory capabilities have captured significant interest from both industrial and academic communities. org) Introduction基于大语言模型（LLM）的智能体最近引起了研究界和工业界的广泛关注。与原始LLM相比，基于LLM的… A Survey on the Memory Mechanism of Large Language Model based Agents Large language model (LLM) based agents have recently attracted much attention from the research and industry communities. Abstract: Large language model (LLM) based agents have recently attracted much attention from the research and industry communities. The synergy between multiple agents and human interaction brings further promise in tackling complex real-world SE problems. The paper reviews previous studies, designs, and evaluations of memory modules for LLM-based agents, and provides code and data for reproducibility. We would like to show you a description here but the site won’t allow us. 13501. In specific, we first discuss “what is” and “why do we need” the memory in LLM-based agents. Paper link: A Survey on Large Language Model based Autonomous Agents Jun 9, 2025 · The memory module functions through three key operations: memory writing, which converts environmental feedback into stored content; memory management, which optimizes information through abstraction, merging, and forgetting; and memory retrieval, which extracts relevant information based on the current context to guide decision-making. It examines how these agents store, retrieve, and utilize information to complete various tasks. Aug 23, 2023 · 📍 This is the first released and published survey paper in the field of LLM-based autonomous agents. It compares textual and parametric memories, emphasizing trade-offs between direct interpretability and efficiency. The survey also highlights challenges and future directions in the field. 4 Mar 15, 2025 · The CoALA framework provides a conceptual understanding of memory in LLM-based agents, distinguishing between working memory and long-term memory. This paper explores the transformative potential of large language model agents in enhancing search and recommendation systems. Apr 21, 2024 · This is a GitHub repository for a survey paper on the memory mechanism of large language model based agents, published on arXiv in 2024. Apr 21, 2024 · This paper reviews previous studies on how to design and evaluate the memory module for LLM-based agents, which are featured in their self-evolving capability. While various memory modules have been proposed for these tasks, the impact of different memory structures across tasks remains insufficiently explored. Dec 17, 2024 · Memory plays a pivotal role in enabling large language model~(LLM)-based agents to engage in complex and long-term interactions, such as question answering (QA) and dialogue systems. org) 这篇论文是由人大高瓴发表的论文，主要是对基于大语言模型的智能体的记忆化机制进行调研。是切入LLM … Jul 25, 2024 · Large Language Model-Based Agents: Leveraged LLMs for reasoning, planning, and interaction, showing promise in diverse applications like software development and scientific research. 最近的相关研究包括：1. Large Language Model (LLM) agents, with goal-driven behaviors and dynamic adaptation capabilities, potentially represent a critical pathway toward artificial general intelligence. To date, LLM-based agents have been applied and shown remarkable 这篇论文名为《A Survey on the Memory Mechanism of Large Language Model based Agents》，作者为Zeyu Zhang等人。 论文的主要目标是对基于大型语言模型（LLM）的智能体的记忆机制进行全面的综述。 随着研究和工业界对LLM智能体的兴趣增长，这种研究越发重要。 Aug 22, 2023 · Recently, through the acquisition of vast amounts of web knowledge, large language models (LLMs) have demonstrated remarkable potential in achieving human-level intelligence. However, current work typically relies on prompt design or fine-tuning strategies applied to vanilla LLMs, which often leads to limited effectiveness or suboptimal performance in complex agent-related 论文链接: [2404. It also presents agent applications, limitations and future directions of the memory mechanism. Dec 31, 2023 · Large language model (LLM) based agents have recently attracted much attention from the research and industry communities. This paper investigates how memory structures and memory Mar 16, 2025 · With the rapid development of Large Language Models (LLMs), LLM-based agents have been widely adopted in various fields, becoming essential for autonomous decision-making and interactive tasks. 《A Survey of Memory in Reinforcement Learning》；2. Jun 9, 2025 · The memory module functions through three key operations: memory writing, which converts environmental feedback into stored content; memory management, which optimizes information through abstraction, merging, and forgetting; and memory retrieval, which extracts relevant information based on the current context to guide decision-making. Nov 24, 2024 · This article presents a comprehensive survey of large language model (LLM)-based autonomous agents, focusing on their construction, applications, and evaluation. The key component to support agent-environment interactions is View recent discussion. 《Memory-Augmented Reinforcement Learning for Robot Navigation》；3. This survey provides a thorough analysis of memory mechanisms essential for LLM-based agents, discussing their evolution and application across interoperable environments: - Defines the importance and functionality of memory in agents - Extensive review of current designs and evaluations of memory modules - Highlights applications where memory plays a pivotal role and suggests future Large language model (LLM) based agents have recently attracted much attention from the research and industry communities. A comprehensive survey on the memory mechanism of LLM-based agents is proposed and many agent applications, where the memory module plays an important role are presented, where the memory module plays an important role. The key component to support agent-environment Abstract Large language model (LLM) based agents have recently attracted much attention from the research and industry communities. Recently, based on the development of using one LLM as a single planning or decision-making agent, LLM-based multi-agent systems have achieved considerable progress in Sep 27, 2024 · In this survey, since this is an interdisciplinary field, we first introduce the background of agent-based modeling and simulation and large language model-empowered agents. This paper provides a comprehensive overview of LLM-based game agents from a holistic viewpoint. For some weeks now, the dblp team has been receiving an exceptionally high number of support and error correction requests from the community. 《Memory-Augmented Monte Carlo Tree Search for General Video Game Playing》等。 Jan 13, 2025 · This survey is the first to systematically summarize the potential techniques for incorporating lifelong learning into LLM-based agents. First, we introduce the conceptual architecture of LLM-based game agents, centered around six essential functional components: perception, memory, thinking, role-playing, action, and learning. 13501] A Survey on the Memory Mechanism of Large Language Model based Agents (arxiv. Compared to standalone LLMs, LLM-based agents substantially extend the versatility and expertise of LLMs by enhancing LLMs with the capabilities of perceiving and utilizing external resources and tools. e. MemGPT and Zep offer concrete implementations of Abstract Large language model (LLM) based agents have recently attracted much attention from the research and industry communities. May 4, 2025 · Abstract. Finally, we illustrate some open problems regarding the memory of current AI systems and outline possible future directions for memory in the era of large language models. The key component to support agent-environment interactions is Jul 11, 2025 · Download Citation | A Survey on the Memory Mechanism of Large Language Model based Agents | Large language model (LLM) based agents have recently attracted much attention from the research and A Survey on the Memory Mechanism of Large Language Model based Agents Zeyu Zhang, Xiaohe Bo, Chen Ma, Rui Li, Xu Chen, Quanyu Dai, Jieming Zhu, Zhenhua Dong, Ji-Rong Wen Jul 7, 2024 · Conclusion The paper presents a novel approach to memory sharing among large language model-based agents, which has the potential to significantly enhance the capabilities and performance of AI systems in a wide range of domains. Apr 8, 2025 · Today, we're excited to share our comprehensive survey paper, "Large Language Model Agent: A Survey on Methodology, Applications and Challenges," which systematically explores this rapidly evolving field. Apr 23, 2024 · TL;DR: This comprehensive survey explores the memory mechanisms of Large Language Model (LLM) based agents, discussing the necessity, design, evaluation, applications, limitations, and future directions of memory in LLM-based agents. Apr 21, 2024 · Large language model (LLM) based agents have recently attracted much attention from the research and industry communities. May 6, 2024 · A Survey on the Memory Mechanism of Large Language Model based Agents（基于大型语言模型的智能体记忆机制调查） 支持智能体与环境交互的关键要素是 智能体的记忆：为了实现人工通用智能（AGI）的最终目标，智能机器应该能够通过自主探索现实世界并从中学习来提高自己 内存的作用： 如何积累知识 处理历史 Apr 21, 2024 · This work designs a time-sharing scheduling strategy, analogous to process scheduling in operating systems, and introduces a hierarchical memory model based on the multi-level cache architecture of operating systems, thereby significantly improving memory retention and retrieval efficiency in LLM - based agents when handling complex tasks. Due to the impressive planning and reasoning abilities of LLMs, they have been used as autonomous agents to do many tasks automatically. 2. The key component to support Mar 14, 2025 · The Rise and Potential of Large Language Model Based Agents: A Survey, arxiv [paper] 💡 A Survey on the Memory Mechanism of Large Language Model based Agents, arxiv [paper] 💡 A Survey on the Memory Mechanism of Large Language Model based Agents Zeyu Zhang, Xiaohe Bo, Chen Ma, Rui Li, Xu Chen, Quanyu Dai, Jieming Zhu, Zhenhua Dong, Ji-Rong Wen. - WooooDyy/LLM-Agent-Paper-List Gaoling School of Artificial Intelligence, Renmin University of China - Cited by 2,301 - LLM-based Agent - Responsible RecSys - Causal Learning A Survey on the Memory Mechanism of Large Language Model based Agents: Paper and Code. In this work, we present a comprehensive and systematic survey on LLM-based agents for SE. [abs], 2024. Abstract—The recent advance in Large Language Models (LLMs) has shaped a new paradigm of AI agents, i. The key component to support agent-environment interactions is To date, LLM-based agents have been applied and shown remarkable effectiveness in Software Engineering (SE). Jul 2, 2025 · To bridge this gap, in this paper, we propose a comprehensive survey on the memory mechanism of LLM-based agents. Compared with original LLMs, LLM-based agents are featured in their self-evolving capability, which is the basis for solving real-world problems that need long-term and complex agent-environment interactions. , LLM-based agents. Recently, large language model based (LLM-based) agents have been widely applied across various fields. Jul 14, 2024 · 本文是LLM系列文章，针对《A Survey on the Memory Mechanism of Large Language Model based Agents》的翻译。 Recent advances in large language models (LLMs) have demonstrated capabilities that surpass human performance in various language-related tasks and exhibit general understanding, reasoning, and decision-making abilities. plex jwfihha vau jjoy lsipy bbkrohe gcaca fqhkzm amfzwk xoyd