RAG & Search
Vector search
Finding items whose vector representations are most similar to a query vector.
Example
A search for “late delivery” also finds a passage describing a shipment that arrived behind schedule.
Why it matters
It helps uncover related material when people use different wording for the same idea.
You'll hear it
“This result sounds similar, but it refers to a different product.”
Can you control it?
Set by the team
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 the closest match always answer the question?
- No. The nearest available item can still be irrelevant if the right material is missing.
- Can it miss an exact code or reference number?
- Yes. Similarity methods may not handle exact identifiers as well as matching their characters directly.
- Does a high score verify the document's claims?
- No. Resemblance to a question is separate from factual reliability.