mlsanityData to 14 Oct 2026

MAGMA: A Multi-Graph based Agentic Memory Architecture for AI Agents

arXiv 2601.03236 ↗First posted 6 Jan 2026cs.AIcited by 52

On the map: LLM agents and tool use → Agent memory systems

Abstract (arXiv)

Memory-Augmented Generation (MAG) extends Large Language Models with external memory to support long-context reasoning, but existing approaches largely rely on semantic similarity over monolithic memory stores, entangling temporal, causal, and entity information. This design limits interpretability and alignment between query intent and retrieved evidence, leading to suboptimal reasoning accuracy. In this paper, we propose MAGMA, a multi-graph agentic memory architecture that represents each memory item across orthogonal semantic, temporal, causal, and entity graphs. MAGMA formulates retrieval as policy-guided traversal over these relational views, enabling query-adaptive selection and structured context construction. By decoupling memory representation from retrieval logic, MAGMA provides transparent reasoning paths and fine-grained control over retrieval. Experiments on LoCoMo and LongMemEval demonstrate that MAGMA consistently outperforms state-of-the-art agentic memory systems in long-horizon reasoning tasks.

Key facts. Cited by 52 papers in the map (data through 14 Oct 2026). Placed in Agent memory systems, one of 16 directions in LLM agents and tool use. Builds on 26 of its 41 references that are in the map. First posted on arXiv 6 Jan 2026.

Builds on

26 of 41 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025Memory in the Age of AI Agents15 Dec
  2. 2025MIRIX: Multi-Agent Memory System for LLM-Based Agents10 Jul
  3. 2025A-MEM: Agentic Memory for LLM Agents17 Feb
  4. 2025Zep: A Temporal Knowledge Graph Architecture for Agent Memory20 Jan
  5. 2024Evaluating Very Long-Term Conversational Memory of LLM Agents27 Feb

Other references · 21

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  3. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  4. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
  5. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  6. 2020Longformer: The Long-Document Transformercited by 3,413
  7. 2017Billion-scale similarity search with GPUscited by 2,497
  8. 2024From Local to Global: A Graph RAG Approach to Query-Focused Summarizationcited by 1,453
  9. 2023The Rise and Potential of Large Language Model Based Agents: A Surveycited by 1,439
  10. 2021Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolationcited by 1,103
  11. 2024LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memorycited by 707
  12. 2023Unifying Large Language Models and Knowledge Graphs: A Roadmapcited by 668
9 more
  1. 2023Causal Reasoning and Large Language Models: Opening a New Frontier for Causalitycited by 353
  2. 2025MemOS: A Memory OS for AI Systemcited by 138
  3. 2024MemoRAG: Boosting Long Context Processing with Global Memory-Enhanced Retrieval Augmentationcited by 119
  4. 2018Sharp Nearby, Fuzzy Far Away: How Neural Language Models Use Contextcited by 113
  5. 2024LongRAG: Enhancing Retrieval-Augmented Generation with Long-context LLMscited by 108
  6. 2023Igniting Language Intelligence: The Hitchhiker's Guide From Chain-of-Thought Reasoning to Language Agentscited by 72
  7. 2024M-RAG: Reinforcing Large Language Model Performance through Retrieval-Augmented Generation with Multiple Partitionscited by 38
  8. 2024Emotional RAG: Enhancing Role-Playing Agents through Emotional Retrievalcited by 26
  9. 2025Cache Mechanism for Agent RAG Systemscited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 52.

52 papers.

Sort:

By area:

  1. 2026Anatomy of Agentic Memory: Taxonomy and Empirical Analysis of Evaluation and System Limitationscited by 21
  2. 2026Governing Evolving Memory in LLM Agents: Risks, Mechanisms, and the Stability and Safety Governed Memory (SSGM) Frameworkcited by 18
  3. 2026GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversationscited by 17
  4. 2026Toward Efficient Agents: Memory, Tool learning, and Planningcited by 16
  5. 2026Jev-Mem: System-One-Controlled Agentic Memory for Efficient AI Agentscited by 14
  6. 2026Bridging Reasoning Trajectories in On-Policy Distillation via Near-Future Guidancecited by 13
  7. 2026Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillationcited by 11
  8. 2026Graph-Native Cognitive Memory for AI Agents: Formal Belief Revision Semantics for Versioned Memory Architecturescited by 9
  9. 2026Escaping the Context Bottleneck: Active Context Curation for LLM Agents via Reinforcement Learningcited by 8
  10. 2026User as Code: Executable Memory for Personalized Agentscited by 7
  11. 2026HyperSkill: Self-Evolving LLM Agents via Hypergraph-Structured Skill Memorycited by 7
  12. 2026HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolutioncited by 6
38 more
  1. 2026The Efficiency Frontier: A Unified Framework for Cost-Performance Optimization in LLM Context Managementcited by 6
  2. 2026ActorMind: Emulating Human Actor Reasoning for Speech Role-Playingcited by 4
  3. 2026Bridging Modalities, Spanning Time: Structured Memory for Ultra-Long Agentic Video Reasoningcited by 4
  4. 2026Goal-Oriented Reasoning for RAG-based Memory in Conversational Agentic LLM Systemscited by 4
  5. 2026V-Mem: Modality-Routed Retrieval for Long-Term Multimodal Agentic Memorycited by 4
  6. 2026ByteRover: Agent-Native Memory Through LLM-Curated Hierarchical Contextcited by 3
  7. 2026FAST: A Synergistic Framework of Attention and State-space Models for Spatiotemporal Traffic Predictioncited by 3
  8. 2026Learning Where to Embed: Noise-Aware Positional Embedding for Query Retrieval in Small-Object Detectioncited by 3
  9. 2026PRISM: Pareto-Efficient Retrieval over Intent-Aware Structured Memory for Long-Horizon Agentscited by 3
  10. 2026MemGuard: Preventing Memory Contamination in Long-Term Memory-Augmented Large Language Modelscited by 3
  11. 2026What Happens Inside Agent Memory? Circuit Analysis from Emergence to Diagnosiscited by 2
  12. 2026PGMem: Tightly Coupled Persona-Memory Graph for Lifelong Personalized Agentscited by 2
  13. 2026VoiceMem: Streaming Dual-Brain Memory for Real-Time Interactioncited by 2
  14. 2026Retrieval-Driven Memory Reconsolidation for Long-Term LLM Agentscited by 2
  15. 2026Memory as Ontology: A Constitutional Memory Architecture for Persistent Digital Citizenscited by 1
  16. 2026Opal: Private Memory for Personal AIcited by 1
  17. 2026Integrating Graphs, Large Language Models, and Agents: Reasoning and Retrievalcited by 1
  18. 2026DarkForest: Less Talk, Higher Accuracy for Multi-Agent LLMscited by 1
  19. 2026AgentSpec: Understanding Embodied Agent Scaffolds Through Controlled Compositioncited by 1
  20. 2026Harness the Memory: A Holistic Evaluation of Memory Substrates in Memory Agentscited by 1
  21. 2026CABLE: Extending the Reach of Memory Retrieval via Complementary Antecedent-Based Linking and Expansioncited by 1
  22. 2026Back to Basics: Let Conversational Agents Remember with Just Retrieval and Generationcited by 0
  23. 2026Self-Aware Vector Embeddings for Retrieval-Augmented Generation: A Neuroscience-Inspired Framework for Temporal, Confidence-Weighted, and Relational Knowledgecited by 0
  24. 2026SemanticAgent: A Semantics-Aware Framework for Text-to-SQL Data Synthesiscited by 0
  25. 2026CogniFold: Always-On Proactive Memory via Cognitive Foldingcited by 0
  26. 2026FundaPod: A Multi-Persona Agent Pod Architecture with Knowledge Graph Memory for AI-Assisted Fundamental Investment Researchcited by 0
  27. 2026Beyond Memory Leaderboards: Evaluating Scientific Memory as Budgeted Context Restorationcited by 0
  28. 2026Towards Trustworthy and Cost-Efficient Data Integration: From Na"ive RAG to Agentic RAGcited by 0
  29. 2026MemPrism: Task-Conditioned Relational Memory Views for Long-Horizon Agentscited by 0
  30. 2026PrimeAgentOrchestrator: Memory-Primed Agent Spawning for Personal AI Infrastructurecited by 0
  31. 2026EvoWiki: Incremental State Overwriting and Traceable Question Answering for Cross-Meeting Knowledge Evolutioncited by 0
  32. 2026Surgical Alignment in Knowledge Graph Training for Clinical Diagnosis with Large Language Modelscited by 0
  33. 2026MEMO: Multimodal Evidence Memory Organization for Long-Horizon LLM Agentscited by 0
  34. 2026Graph-Based Personalized Memory for LLM Agents: Representation, Evolution, Retrieval, and Evaluationcited by 0
  35. 2026Propose, Verify, Commit: Evidence-Grounded Memory for Long-Horizon Multi-Actor Conversationscited by 0
  36. 2026Contract Memory Compiler: Resolve, Then Traversecited by 0
  37. 2026Remember Before You're Asked: MemDream for Self-Probing Memory Evolutioncited by 0
  38. 2026TAGGRAPH: Tag-Augmented Graphs for Graph Retrieval of Agent Persistent Historiescited by 0

50 most cited of 52 All 52 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Memory-Orchestrated Semantic System (MOSS): An Auditable Agentic Memory Architecture
  2. 2026Memanto: Typed Semantic Memory with Information-Theoretic Retrieval for Long-Horizon Agents
  3. 2026Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents
  4. 2026MemWeaver: Weaving Hybrid Memories for Traceable Long-Horizon Agentic Reasoning
  5. 2025LiCoMemory: Lightweight and Cognitive Agentic Memory for Efficient Long-Term Reasoning
  6. 2025Decentralizing AI Memory: SHIMI, a Semantic Hierarchical Memory Index for Scalable Agent Reasoning
  7. 2026AMA: Adaptive Memory via Multi-Agent Collaboration
  8. 2025Hierarchical Memory for High-Efficiency Long-Term Reasoning in LLM Agents