Embedding vector Machine Learning
noun phrase
Definition: A dense numerical vector produced by an embedding model/layer to represent an input item (e.g., a word, sentence, image, or user) [Google ML Glossary].
Example in context: “For each review we produce an embedding vector, by extracting the vector associated with the special [CLS] token from the last hidden layer of BERT.” [Apel et al. 2022]
Synonym: embedding (context-dependent)
Related terms: feature vector, latent vector