{"paper":{"title":"A Survey on the Memory Mechanism of Large Language Model based Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Memory mechanisms let LLM-based agents handle long-term interactions by storing and retrieving information beyond single prompts.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chen Ma, Jieming Zhu, Ji-Rong Wen, Quanyu Dai, Rui Li, Xiaohe Bo, Xu Chen, Zeyu Zhang, Zhenhua Dong","submitted_at":"2024-04-21T01:49:46Z","abstract_excerpt":"Large language model (LLM) based agents have recently attracted much attention from the research and industry communities. 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. The key component to support agent-environment interactions is the memory of the agents. While previous studies have proposed many promising memory mechanisms, they are scattered in different papers, and there lacks a systematical review to summarize and compare these"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Previous studies have proposed many promising memory mechanisms, they are scattered in different papers, and there lacks a systematical review to summarize and compare these works from a holistic perspective, failing to abstract common and effective designing patterns for inspiring future studies.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the reviewed papers are representative of the field and that the proposed categorization successfully abstracts common designing patterns that will guide future work.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A systematic review of memory designs, evaluation methods, applications, limitations, and future directions for LLM-based agents.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Memory mechanisms let LLM-based agents handle long-term interactions by storing and retrieving information beyond single prompts.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"17059ff75528188666eab42f6ef5a7efe773cc48216a49163e86aaf7d52d49fc"},"source":{"id":"2404.13501","kind":"arxiv","version":1},"verdict":{"id":"21821099-2c5b-4672-be8e-4a844f2d80f2","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T07:15:24.951273Z","strongest_claim":"Previous studies have proposed many promising memory mechanisms, they are scattered in different papers, and there lacks a systematical review to summarize and compare these works from a holistic perspective, failing to abstract common and effective designing patterns for inspiring future studies.","one_line_summary":"A systematic review of memory designs, evaluation methods, applications, limitations, and future directions for LLM-based agents.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the reviewed papers are representative of the field and that the proposed categorization successfully abstracts common designing patterns that will guide future work.","pith_extraction_headline":"Memory mechanisms let LLM-based agents handle long-term interactions by storing and retrieving information beyond single prompts."},"references":{"count":173,"sample":[{"doi":"","year":2023,"title":"ChatDev: Communicative Agents for Software Development","work_id":"5d8a3650-ab78-4991-b0d3-5309b59c690f","ref_index":1,"cited_arxiv_id":"2307.07924","is_internal_anchor":true},{"doi":"","year":2023,"title":"S$^3$: Social-network Simulation System with Large Language Model-Empowered Agents","work_id":"ae6d9bb2-30d7-4b8c-8140-bcb69cc5c24c","ref_index":2,"cited_arxiv_id":"2307.14984","is_internal_anchor":true},{"doi":"","year":2023,"title":"A Survey on Large Language Model based Autonomous Agents","work_id":"47f7e8a3-3732-4530-b412-d9c984ce99ed","ref_index":3,"cited_arxiv_id":"2308.11432","is_internal_anchor":true},{"doi":"","year":2023,"title":"The Rise and Potential of Large Language Model Based Agents: A Survey","work_id":"985ca219-7e34-4c4f-bdc5-ccd39763ad61","ref_index":4,"cited_arxiv_id":"2309.07864","is_internal_anchor":true},{"doi":"","year":2023,"title":"Reflexion: Language agents with verbal reinforcement learning","work_id":"08870c88-ca3e-4168-a457-43a20fda92d3","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":173,"snapshot_sha256":"a4c4a8bf9ca06334acd763512ce92d0745c7a9427cfa0c78b04fc70834aaff4a","internal_anchors":30},"formal_canon":{"evidence_count":1,"snapshot_sha256":"3568d591f98759d1ce1bde10c56b79588be8a7c0c0c294d6ddd7107622d3950c"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}