Technical
Embedding
A dense vector representation of text in a high-dimensional space where semantic similarity corresponds to geometric proximity. Used for search, clustering, and recommendation.
Explained at five levels
Level 1
Turning words into secret number codes so the computer can understand how similar or different things are.
Level 2
A way of converting text into numbers that capture meaning — similar words get similar numbers, so the AI can understand relationships.
Level 3
A dense vector representation of text in a high-dimensional space where semantic similarity corresponds to geometric proximity. Used for search, clustering, and recommendation.
Level 4
A learned mapping from discrete tokens or passages to continuous vectors in ℝⁿ, where cosine similarity approximates semantic relatedness. Fundamental to vector search, RAG retrieval, and representation learning.
Level 5
A function f: X → ℝⁿ mapping inputs to a metric space where the distance function approximates a target similarity — trained via contrastive loss, producing representations useful for nearest-neighbor retrieval, clustering, and downstream transfer.
Definitions are educational summaries. Terminology can vary by source and context.
Sources
- NIST Trustworthy and Responsible AI Resource Center: Glossary — National Institute of Standards and Technology. Accessed 2026-07-20.