Retrieval & ranking
Passage Embedding
A numeric vector that captures a passage's meaning so similar passages sit near each other.
A passage embedding is a numeric vector that represents the meaning of a passage, generated by an embedding model. Passages with similar meaning produce vectors that sit close together, which is how a retrieval system finds content relevant to a query even when the exact words differ.
Embeddings are why keyword matching is no longer the whole story. An engine can retrieve your passage for a query that shares no words with it, as long as the meanings align. This rewards writing that clearly expresses a concept over writing that merely repeats a keyword.
You do not create embeddings yourself for third-party engines, but understanding them explains the goal: express each idea clearly and specifically so its meaning is unambiguous when embedded and matched.
Related terms
Vector Embedding
A list of numbers that encodes meaning so related content sits close in vector space.
Semantic Similarity
How close two pieces of text are in meaning, measured by comparing their embeddings.
Cosine Similarity
A score from -1 to 1 measuring how closely two vectors point in the same direction.
Chunk Retrieval
Fetching the top few most relevant text passages from a vector index for a query.
Chunking
Splitting a document into small passages an AI can retrieve and quote one at a time.
Want to know where you stand in AI answers?
We’ll show you which engines cite you, and which don’t.
Get a free AI citation audit