Skip to content

RAG & Search

Cross-encoder

AI that examines two pieces of information together to judge how closely they are related.

Example

A reranker reads a query together with a candidate passage.

Why people use it

It can compare a question and a possible answer passage in greater detail.

What you'll hear

“Read the question and passage together before ranking them.”

What this means for you

Use the added detail where it justifies the computing work.

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

Can this rescue a useful document that was never found?
Not by comparing a shortlist alone. A missing document must first reach the set of items being compared.
Why can it cost more to search this way?
Each pair needs to be processed together, which adds work when there are many candidate passages.
Can it be used after a faster search?
Yes. A fast first search can narrow the choices before a more detailed comparison.

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