Defined scope
- Information retrieval
- Embeddings and indexes
- Context assembly
- Source grounding
- Citation
- Evaluation
CONCEPT
Retrieval-Augmented Generation
Retrieval-augmented generation is an architecture that retrieves external information at query time and supplies selected evidence to a generative model before or during response production.
RAG connects search, embeddings, document segmentation, metadata, provenance, context assembly, language models and answer evaluation.
active
research
EDITORIAL FRAME
What this entry establishes.
A concise view of its scope, position, limitations and supporting sources.
Editorial position
- RAG separates updatable external knowledge from model parameters.
- Retrieval quality, chunk design, metadata and answer attribution determine system reliability.
Explicit limits
- Adding an entire uncontrolled document dump to a prompt
- Assuming retrieval guarantees that generated claims are supported
Tags and disciplines
RAGRetrievalEmbeddingsVector SearchGroundingArtificial IntelligenceKnowledge SystemsInformation ArchitectureMachine Learning
References behind this page
- Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks Patrick Lewis et al. / arXiv / 2020-05-22
Definition
Retrieval-augmented generation combines a retriever with a generative model. The retriever selects passages, records or graph neighborhoods relevant to a request. The model receives that material as context and produces a response.
Architecture
A practical system includes ingestion, segmentation, metadata, indexing, query transformation, retrieval, ranking, context assembly, generation, citation and evaluation. Each stage can fail independently.
Knowledge graph role
Vector similarity can find semantically related passages, while graph relations can preserve identity, type, provenance and explicit paths. Hybrid retrieval can use both rather than treating embeddings as a replacement for structured knowledge.
Electronic Artefacts position
The Knowledge Hub can provide high-quality retrieval material because its pages expose stable entities, sources, confidence labels and relations. Future RAG systems should cite canonical records and retain the distinction between retrieved evidence and generated synthesis.
Limitations
Retrieved passages may be irrelevant, outdated, inaccessible or misleading. A model can ignore evidence or attach a citation to an unsupported claim. Evaluation must test retrieval recall, source precision, faithfulness and answer usefulness separately.
References
See Lewis et al., Knowledge Graph, Metadata, Provenance and Large Language Model.
DOCUMENTED RELATIONSHIPS
Connected work and ideas.
Each link names the relationship between two entries and why it matters.
AI Search, Structured Content and Knowledge Graph SEO
AI Search, Structured Content and Knowledge Graph SEO connects AI search visibility to retrieval-augmented generation patterns.
Retrieval-Augmented Generation and Knowledge Systems
Retrieval-Augmented Generation and Knowledge Systems documents RAG architecture end to end.
Prompt Injection and Trust Boundaries in AI Knowledge Systems
Prompt Injection and Trust Boundaries in AI Knowledge Systems documents Retrieval-Augmented Generation as one of its declared subjects.
Knowledge Hub Third Wave
Retrieval-Augmented Generation is an explicit member of the Knowledge Hub Third Wave collection.
Record details Metadata, sharing and citation
Retrieval-Augmented Generation
Cite this page
Retrieval-Augmented Generation. 1.0.0. Electronic Artefacts, 2026-06-24. https://electronicartefacts.com/knowledge/concepts/retrieval-augmented-generation/
Nearby relationships
4 public links connect this page to nearby projects, concepts and references.