Embedding Machine Learning
noun
Definition: A numerical vector representation of data such as text, images, audio, or video that captures relationships among inputs in a continuous vector space. Embeddings are used to encode similarity, semantic relatedness, and other latent structure in a form suitable for machine learning models [Google ML Glossary].
Example in context: “The challenge for text applications is to develop text embeddings that are semantically rich…” [Muennighoff et al. 2025]
Synonyms: vector representation; dense representation
Related terms: embedding vector; sentence embedding; multimodal embedding; semantic representation