Agentic AI Reference Architecture

Cognition · Software component

Factuality Verifier

Software componentCognitionCognition & MemoryVariation point (abstract)arc:FactualityVerifier

A semantic validation component that fact-checks generated claims against knowledge bases or retrieved evidence, producing correctness signals for runtime validation or reward computation.

Responsibility. Verifies factual correctness of generated content against trusted knowledge.

Also known as: Semantic validation layer, Fact-checking layer, Factuality check

guardsinvokesis specialized byis invoked byReasoning Engine: guardsReasoning EngineRetriever: invokesRetrieverFact Checking Rail: is specialized byFact Checking RailComposite Reward Scorer: is invoked byComposite Reward Scorer
Direct neighbourhood (hover for relationship types)

Variants

VariantWhen to choose
Fact Checking Rail abstract—

Relationships

invokes dependency

is invoked by dependency

guards control

Design guidance

Classification

Patterns
Grounded validationVerifiable correctness signal
Quality attributes
Functional suitability: correctness and validity (ISO/IEC 25010 | NIST AI RMF: valid)Interaction capability (ISO/IEC 25010)
Risks mitigated
HallucinationHallucinations cascading across multi-step reasoningPreference for fluent but incorrect responses

Sources

  1. Ch3.5: T. Nguyen, "Prompt Optimization, Few-Shot Learning, Fine-Tuning," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 3.5. ISBN: 9798244538229.
  2. Ch7.1A: T. Nguyen, "Advanced Implementation with Nvidia NEMO Framework and Nvlink," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 7.1A. ISBN: 9798244538229.