Concept · How AI answers work
Retrieval-augmented generation
Definition
Fetching documents at question time and giving them to a language model to write its answer from.
What it means in practice
Usually shortened to RAG. It explains why AI answers change without any model being retrained: the retrieval step ran again and returned a different set of pages. It also explains why being cited is partly a retrieval problem rather than a writing problem, and why the same brand can appear in one engine and not another that retrieved from a different index.
How refd treats it
This is category vocabulary, not a refd metric. refd measures what AI answers actually said across repeated runs, and publishes the raw answer behind every number. Where a term above carries a claim about what causes visibility, refd treats that claim as a hypothesis to test rather than a setting to configure.
This definition is editorial. It describes how the term is used across the field rather than something the software computes, so unlike the metric pages it is not generated from the product.
See your evidence