Embedding

A dense vector that places text in a space where distance means semantic similarity. Retrieval compares a query vector against stored passage vectors, which is why a vector database is the storage layer for RAG.

Why exams ask this

Tested on what embeddings do NOT give you: they carry no notion of recency, authority, or truth. Similarity is not correctness.

Relevant to

Related concepts

Resources

No resources linked to this concept yet.