mlsanityData to 14 Oct 2026

Contextual Agentic Memory is a Memo, Not True Memory

arXiv 2604.27707 ↗First posted 30 Apr 2026cs.AI cs.CLcited by 6

On the map: Retrieval-augmented generation → Agentic RAG, agents and memory

also in 1 more
Abstract (arXiv)

Current agentic memory systems (vector stores, retrieval-augmented generation, scratchpads, and context-window management) do not implement memory: they implement lookup. We argue that treating lookup as memory is a category error with provable consequences for agent capability, long-term learning, and security. Retrieval generalizes by similarity to stored cases; weight-based memory generalizes by applying abstract rules to inputs never seen before. Conflating the two produces agents that accumulate notes indefinitely without developing expertise, face a provable generalization ceiling on compositionally novel tasks that no increase in context size or retrieval quality can overcome, and are structurally vulnerable to persistent memory poisoning as injected content propagates across all future sessions. Drawing on Complementary Learning Systems theory from neuroscience, we show that biological intelligence solved this problem by pairing fast hippocampal exemplar storage with slow neocortical weight consolidation, and that current AI agents implement only the first half. We formalize these limitations, address four alternative views, and close with a co-existence proposal and a call to action for system builders, benchmark designers, and the memory community.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Placed in Agentic RAG, agents and memory, one of 15 directions in Retrieval-augmented generation. Builds on 49 of its 67 references that are in the map. First posted on arXiv 30 Apr 2026.

Builds on

49 of 67 references.

References · 49

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  2. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  3. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  4. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
  5. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
  6. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  7. 2022Locating and Editing Factual Associations in GPTcited by 2,691
  8. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,191
  9. 2023Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injectioncited by 1,443
  10. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,407
  11. 2020Transformer Feed-Forward Layers Are Key-Value Memoriescited by 1,336
  12. 2021Improving language models by retrieving from trillions of tokenscited by 1,293
37 more
  1. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,063
  2. 2022Mass-Editing Memory in a Transformercited by 1,001
  3. 2019Generalization through Memorization: Nearest Neighbor Language Modelscited by 800
  4. 2025Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memorycited by 779
  5. 2022What learning algorithm is in-context learning? Investigations with linear modelscited by 684
  6. 2024A Survey on the Memory Mechanism of Large Language Model based Agentscited by 634
  7. 2023MemoryBank: Enhancing Large Language Models with Long-Term Memorycited by 556
  8. 2024InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agentscited by 538
  9. 2024Learning to (Learn at Test Time): RNNs with Expressive Hidden Statescited by 355
  10. 2025Memory in the Age of AI Agentscited by 303
  11. 2024PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Modelscited by 267
  12. 2020COGS: A Compositional Generalization Challenge Based on Semantic Interpretationcited by 263
  13. 2023Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMscited by 178
  14. 2025Memory Injection Attacks on LLM Agents via Query-Only Interactioncited by 144
  15. 2023Few-shot Fine-tuning vs. In-context Learning: A Fair Comparison and Evaluationcited by 136
  16. 2026MemRL: Self-Evolving Agents via Runtime Reinforcement Learning on Episodic Memorycited by 129
  17. 2024Mitigating Catastrophic Forgetting in Large Language Models with Self-Synthesized Rehearsalcited by 118
  18. 2022Compositional Semantic Parsing with Large Language Modelscited by 95
  19. 2026Skill-SD: Skill-Conditioned Self-Distillation for Multi-turn LLM Agentscited by 90
  20. 2022Continual Learning of Natural Language Processing Tasks: A Surveycited by 83
  21. 2025MemOS: An Operating System for Memory-Augmented Generation (MAG) in Large Language Modelscited by 73
  22. 2025Nested Learning: The Illusion of Deep Learning Architecturescited by 72
  23. 2024Knowledge Circuits in Pretrained Transformerscited by 60
  24. 2023Skills-in-Context Prompting: Unlocking Compositionality in Large Language Modelscited by 33
  25. 2026Experience Compression Spectrum: Unifying Memory, Skills, and Rules in LLM Agentscited by 17
  26. 2025One Surrogate to Fool Them All: Universal, Transferable, and Targeted Adversarial Attacks with CLIPcited by 15
  27. 2025Latent learning: episodic memory complements parametric learning by enabling flexible reuse of experiencescited by 15
  28. 2025Context Is What You Need: The Maximum Effective Context Window for Real World Limits of LLMscited by 12
  29. 2026Memory in the LLM Era: Modular Architectures and Strategies in a Unified Frameworkcited by 12
  30. 2025Task Generalization With AutoRegressive Compositional Structure: Can Learning From D Tasks Generalize to DT Tasks?cited by 11
  31. 2025CLIP-Guided Backdoor Defense through Entropy-Based Poisoned Dataset Separationcited by 10
  32. 2025Analyzing the Effects of Supervised Fine-Tuning on Model Knowledge from Token and Parameter Levelscited by 7
  33. 2025Breaking the Stealth-Potency Trade-off in Clean-Image Backdoors with Generative Trigger Optimizationcited by 7
  34. 2026From Multi-Agent to Single-Agent: When Is Skill Distillation Beneficial?cited by 6
  35. 2026Self-Consolidation for Self-Evolving Agentscited by 5
  36. 2026ParamMem: Augmenting Language Agents with Parametric Reflective Memorycited by 5
  37. 2026Fine-Tuning vs. RAG for Multi-Hop Question Answering with Novel Knowledgecited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 6.

6 papers.

  1. 2026Learning the Context of Errors: Black-Box Online Adaptation of Time Series Foundation Modelscited by 9
  2. 2026When Agent Automation Becomes Profitable: Quantifying and Insuring Autonomous AI Risk through Trace-Economic Underwritingcited by 3
  3. 2026LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via State Proprioceptioncited by 3
  4. 2026MemoryCPT: An End-to-End Agent Memory Framework for Cost-Performance Trade-offcited by 1
  5. 2026One Rollout Is All You Get: Fully Test-Time Adaptation for GUI Agentscited by 0
  6. 2026EmbodiedCLBench: Evaluating Continual Learning for Self-Evolving Embodied Agentscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025AI Meets Brain: Memory Systems from Cognitive Neuroscience to Autonomous Agents
  2. 2026Hippocampus: An Efficient and Scalable Memory Module for Agentic AI
  3. 2026Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents
  4. 2026A Survey of Agent Memory in the Second Half: Towards Self-Evolving and Long-Horizon Agents
  5. 2026Exploring Cross-Scenario Generality of Agentic Memory Systems: Diagnostics and a Strong Baseline
  6. 2026The AI Hippocampus: How Far are We From Human Memory?
  7. 2026RaMem: Contextual Reinstatement for Long-term Agentic Memory
  8. 2026Memory as Middleware for Self-Improving AI Agents