Sentence Embedding
A dense mathematical vector representation of an entire sentence or paragraph, capturing its full semantic meaning in a fixed-dimensional space.
Think of It Like This
Like boiling down a 500-page novel into a single, highly dense summary paragraph that captures the entire core plot.
Unlike word embeddings which represent single tokens, sentence embeddings use architectures like SBERT (Sentence-BERT) to pool token vectors into a unified representation. They are the fundamental building blocks for semantic search, document clustering, and retrieval-augmented generation (RAG) pipelines.