mlsanityData to 7 Oct 2026

Rethinking Memory in LLM based Agents: Representations, Operations, and Emerging Topics

arXiv 2505.00675 ↗First posted 1 May 2025cs.CLcited by 36

On the map: LLM reasoning → Reasoning for agents, planning and code

also in 1 more
Abstract (arXiv)

Memory is fundamental to large language model (LLM)-based agents, but existing surveys emphasize application-level use (e.g., personalized dialogue), while overlooking the atomic operations governing memory dynamics. This work categorizes memory into parametric (implicit in model weights) and contextual (explicit external data, structured/unstructured) forms, and defines six core operations: Consolidation, Updating, Indexing, Forgetting, Retrieval, and Condensation. Mapping these dimensions reveals four key research topics: long-term, long-context, parametric modification, and multi-source memory. The taxonomy provides a structured view of memory-related research, benchmarks, and tools, clarifying functional interactions in LLM-based agents and guiding future advancements. The datasets, papers, and tools are publicly available at https://github.com/Elvin-Yiming-Du/Survey_Memory_in_AI.

Key facts. Cited by 36 papers in the map (data through 7 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 89 of its 383 references that are in the map. First posted on arXiv 1 May 2025.

Builds on

89 of 383 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Enhancing Reasoning with Collaboration and Memory7 Mar

Other references · 88

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  3. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,179
  4. 2023PaLM-E: An Embodied Multimodal Language Modelcited by 2,037
  5. 2020Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Stepscited by 1,259
  6. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,203
  7. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,000
  8. 2022Mass-Editing Memory in a Transformercited by 976
  9. 2024Evaluating Very Long-Term Conversational Memory of LLM Agentscited by 800
  10. 2024DeepSeek-V3 Technical Reportcited by 724
  11. 2023CogAgent: A Visual Language Model for GUI Agentscited by 692
  12. 2024LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memorycited by 656
38 more
  1. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 655
  2. 2024A Survey on the Memory Mechanism of Large Language Model based Agentscited by 622
  3. 2025UI-TARS: Pioneering Automated GUI Interaction with Native Agentscited by 584
  4. 2023LaMP: When Large Language Models Meet Personalizationcited by 425
  5. 2025Zep: A Temporal Knowledge Graph Architecture for Agent Memorycited by 372
  6. 2024The Faiss librarycited by 370
  7. 2024LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitaskscited by 367
  8. 2024Titans: Learning to Memorize at Test Timecited by 303
  9. 2024LongRoPE: Extending LLM Context Window Beyond 2 Million Tokenscited by 302
  10. 2024Knowledge Conflicts for LLMs: A Surveycited by 274
  11. 2021Beyond Goldfish Memory: Long-Term Open-Domain Conversationcited by 266
  12. 2024MUSE: Machine Unlearning Six-Way Evaluation for Language Modelscited by 251
  13. 2025From RAG to Memory: Non-Parametric Continual Learning for Large Language Modelscited by 223
  14. 2022Memorizing Transformerscited by 219
  15. 2023In-Context Unlearning: Language Models as Few Shot Unlearnerscited by 214
  16. 2023LongNet: Scaling Transformers to 1,000,000,000 Tokenscited by 201
  17. 2023Synapse: Trajectory-as-Exemplar Prompting with Memory for Computer Controlcited by 180
  18. 2024A Comprehensive Study of Knowledge Editing for Large Language Modelscited by 157
  19. 2025WorldMem: Long-term Consistent World Simulation with Memorycited by 152
  20. 2023ChatDB: Augmenting LLMs with Databases as Their Symbolic Memorycited by 136
  21. 2024xRAG: Extreme Context Compression for Retrieval-augmented Generation with One Tokencited by 123
  22. 2023Chain-of-Knowledge: Grounding Large Language Models via Dynamic Knowledge Adapting over Heterogeneous Sourcescited by 121
  23. 2024Hello Again! LLM-powered Personalized Agent for Long-term Dialoguecited by 121
  24. 2024MemoRAG: Boosting Long Context Processing with Global Memory-Enhanced Retrieval Augmentationcited by 118
  25. 2024RetrievalAttention: Accelerating Long-Context LLM Inference via Vector Retrievalcited by 114
  26. 2022Long Time No See! Open-Domain Conversation with Long-Term Persona Memorycited by 107
  27. 2023SCM: Enhancing Large Language Model with Self-Controlled Memory Frameworkcited by 98
  28. 2023KGA: A General Machine Unlearning Framework Based on Knowledge Gap Alignmentcited by 94
  29. 2023MemoChat: Tuning LLMs to Use Memos for Consistent Long-Range Open-Domain Conversationcited by 94
  30. 2025MemInsight: Autonomous Memory Augmentation for LLM Agentscited by 84
  31. 2024Optimus-1: Hybrid Multimodal Memory Empowered Agents Excel in Long-Horizon Taskscited by 83
  32. 2024Embodied LLM Agents Learn to Cooperate in Organized Teamscited by 81
  33. 2022Memory Replay with Data Compression for Continual Learningcited by 80
  34. 2023NExT-Chat: An LMM for Chat, Detection and Segmentationcited by 79
  35. 2023Advancing Transformer Architecture in Long-Context Large Language Models: A Comprehensive Surveycited by 71
  36. 2024Prompt Compression for Large Language Models: A Surveycited by 65
  37. 2025A Survey of Personalized Large Language Models: Progress and Future Directionscited by 65
  38. 2024Human-inspired Episodic Memory for Infinite Context LLMscited by 62

50 most cited of 88 All 88 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 14; 2026: 22.

36 papers.

Sort:

By area:

  1. 2025Memory in the Age of AI Agentscited by 285
  2. 2025AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challengescited by 167
  3. 2025MemOS: A Memory OS for AI Systemcited by 129
  4. 2025MemOS: An Operating System for Memory-Augmented Generation (MAG) in Large Language Modelscited by 68
  5. 2026A Survey of Agent Memory in the Second Half: Towards Self-Evolving and Long-Horizon Agentscited by 40
  6. 2025Look Back to Reason Forward: Revisitable Memory for Long-Context LLM Agentscited by 37
  7. 2025StoryMem: Multi-shot Long Video Storytelling with Memorycited by 35
  8. 2025MMLongBench: Benchmarking Long-Context Vision-Language Models Effectively and Thoroughlycited by 29
  9. 2025MemR3: Memory Retrieval via Reflective Reasoning for LLM Agentscited by 17
  10. 2026Self-Improvements in Modern Agentic Systems: A Surveycited by 17
  11. 2026MemOCR: Layout-Aware Visual Memory for Efficient Long-Horizon Reasoningcited by 15
  12. 2026Memory in the LLM Era: Modular Architectures and Strategies in a Unified Frameworkcited by 12
24 more
  1. 2025Memory-T1: Reinforcement Learning for Temporal Reasoning in Multi-session Agentscited by 11
  2. 2026Toward Efficient Agents: Memory, Tool learning, and Planningcited by 11
  3. 2025Memory-Augmented Transformers: A Systematic Review from Neuroscience Principles to Enhanced Model Architecturescited by 9
  4. 2026The Vision Wormhole: Latent-Space Communication in Heterogeneous Multi-Agent Systemscited by 8
  5. 2026Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Applicationcited by 8
  6. 2025Knowledge Augmented Complex Problem Solving with Large Language Models: A Surveycited by 6
  7. 2026LMEB: Long-horizon Memory Embedding Benchmarkcited by 6
  8. 2025Smarter Together: Creating Agentic Communities of Practice through Shared Experiential Learningcited by 5
  9. 2025A Scenario-Driven Cognitive Approach to Next-Generation AI Memorycited by 3
  10. 2026MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Modelscited by 3
  11. 2025Contextual Memory Intelligence -- A Foundational Paradigm for Human-AI Collaboration and Reflective Generative AI Systemscited by 2
  12. 2026Memory for Large Language Modelscited by 2
  13. 2026Not All Thoughts Need HBM: Semantics-Aware Memory Hierarchy for LLM Reasoningcited by 1
  14. 2026SF-AMS: Strategic Forgetting for Structured Memory in LLM Agentcited by 1
  15. 2026Learning Design Skills as Memory Policies for Agentic Photonic Inverse Designcited by 0
  16. 2026APeB: Benchmarking Personalization Ability of Large Language Model Agentscited by 0
  17. 2026Towards a Formal Definition of Agent Memory: Basis, Span, Optimality, and the Sequential Memory Problemcited by 0
  18. 2026MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Usecited by 0
  19. 2026UTILMEM: Benchmarking Evidence Utilization in Long-Term Conversational Memorycited by 0
  20. 2026Where to Look and What to Use: Retrieve-Localize-Generate for Long-Term Conversational Memory Question Answeringcited by 0
  21. 2026Personalizing LLM Agent Memory Using Biometricscited by 0
  22. 2026Complementary Roles of Activation and Parametric Memory in Few-Shot Learningcited by 0
  23. 2026LSTMem: Hierarchical Long Short-Term Online Memory for Large Language Modelscited by 0
  24. 2026DISCO: Distributed Long Context Scaling with Grounding-Reasoning Disaggregationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontiers
  2. 2025Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions
  3. 2024On the Structural Memory of LLM Agents
  4. 2026Graph-Based Personalized Memory for LLM Agents: Representation, Evolution, Retrieval, and Evaluation
  5. 2026Beyond Dialogue Time: Temporal Semantic Memory for Personalized LLM Agents
  6. 2026Evaluating Memory Structure in LLM Agents
  7. 2023An In-depth Survey of Large Language Model-based Artificial Intelligence Agents
  8. 2026Anatomy of Agentic Memory: Taxonomy and Empirical Analysis of Evaluation and System Limitations