Memory · Data store

Semantic Memory Store

Data storeMemoryCognition & Memoryarc:SemanticMemoryStore

Long-term memory of facts, concepts and structured user/domain knowledge, searched by embedding similarity.

Responsibility. Stores distilled facts and concepts for similarity retrieval.

Also known as: Semantic memory, Long-term memory (facts), Long-term memory, Abstract domain knowledge, Long-term user preference memory

Direct neighbourhood (hover for relationship types)

Relationships

is read by dependency

is written by dependency

is guarded by control

Memory Write Validator Ch1.4

is audited by assurance

Knowledge Base Auditor Ch5.8

Design guidance

SHOULD be kept separate from episodic and procedural memory to avoid retrieval noise.
SHOULD receive general patterns abstracted from many episodes so that specific episodes can later be forgotten while the pattern persists.
SHOULD choose the representation (vector, knowledge graph or hybrid) by how the agent must reason, not by which technology is better.
MUST tag knowledge with temporal validity so historical facts are not mixed with current facts.
SHOULD weight retrieval by source reliability rather than treating all ingested sources equally.

Quantitative guidance

As stated by the sources; verify before use.

Systems without long-term memory drop 15-30% accuracy on personalization tasks (Ch3.7).
A ~20-token semantic entry can replace ~5,000 tokens of source papers (Ch5.9).

Classification

Patterns: Vector similarity searchCross-modal semantic anchorsTiered memoryRetrieval-augmented grounding over external factual knowledgeVector, knowledge-graph or hybrid representationTemporal versioning with validity periodsConfidence-weighted factsDistilled, structured facts for token-efficient augmentationEntity-relation-observation knowledge graph (Ref5.05)Hierarchical memory architecturePersonalization
Technologies: QdrantPineconeChromaVector storeModel Context Protocol memory server
Quality attributes: Functional suitability: correctness and validity (ISO/IEC 25010 | NIST AI RMF: valid)Interaction capability (ISO/IEC 25010)Explainability (NIST AI RMF: explainable and interpretable)
Risks mitigated: Stale parametric knowledge (training cutoff)Missing proprietary organizational knowledgeHallucinated facts

Sources

Ch1.4: T. Nguyen, "Memory and Perception Systems," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 1.4. ISBN: 9798244538229.
Ch1.7B: T. Nguyen, "Relational Reasoning with Knowledge Graphs - Hybrid RAG+KG Integration," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 1.7B. ISBN: 9798244538229.
Ch3.7: T. Nguyen, "Tool Usage Auditing," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 3.7. ISBN: 9798244538229.
Ch5.1: T. Nguyen, "Chain-of-Thought (CoT) Fundamentals," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.1. ISBN: 9798244538229.
Ch5.2: T. Nguyen, "Tree-of-Thought (ToT) Fundamentals," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.2. ISBN: 9798244538229.
Ch5.7: T. Nguyen, "Episodic Memory," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.7. ISBN: 9798244538229.
Ch5.8: T. Nguyen, "Semantic Memory," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.8. ISBN: 9798244538229.
Ch5.9: T. Nguyen, "Working Memory," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.9. ISBN: 9798244538229.
Ch10.1: T. Nguyen, "Conversational UI," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 10.1. ISBN: 9798244538229.
Ref5.04: C. Stryker, "What is AI agent memory?," IBM Think. Accessed: Sep. 27, 2026. [Online]. Available: https://www.ibm.com/think/topics/ai-agent-memory
Ref5.05: Model Context Protocol, "Model Context Protocol servers," GitHub repository. Accessed: Sep. 27, 2026. [Online]. Available: https://github.com/modelcontextprotocol/servers