Skip to content

RAG & Search

GraphRAG

Graph retrieval-augmented generation

AI answering questions using information organized as connected people, things and relationships.

Example

A system connects people or things and relationships before answering a broad document question.

Why people use it

It helps connect information spread across related people, places or topics.

What you'll hear

“Use the relationships between these documents to answer the question.”

What this means for you

Validate the graph and compare results on representative questions.

Can you control it?

Developer-only

The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.

Common questions

Does using a graph guarantee more accurate answers?
No. Mistakes or missing connections in the stored information can carry through into answers.
Can it help with a question spanning many documents?
It can, when the stored relationships capture the connections the question needs.
Can building the graph add work?
Yes. Extracting, checking and updating relationships can take additional time and computing resources.

Related terms

Still have questions?

Up to 500 characters.

Ask LATHIC about AI. Relevant glossary entries may be included.

Your question, the glossary entries it matches, and a rotating pseudonymous identifier go to Microsoft Azure’s OpenAI service through Vercel AI Gateway to generate an answer. Zero retention and no training are required of the provider, and LATHIC does not save your question or answer. Privacy Notice