Embedding
Also: embeddings
A list of numbers used to compare items such as texts. Texts with similar meanings may have similar lists.
An embedding turns a piece of text into a position on a map. Texts that mean similar things land close together, even with different words. Searching by meaning means looking for the nearest neighbours on the map.
The fuller definition
An embedding is a numerical representation of an item, such as text or an image, in a vector space. Similarity between representations can support search and comparison.
Go deeper
Related phrasing can be nearby in an embedding space. Real embeddings have many dimensions, and the model determines similarity. Nearby does not mean legally relevant or factually correct.
A concrete example
An embedding model places related descriptions closer together under its chosen similarity measure.
The distinction that matters
Nearness is model-dependent similarity, not a probability of legal relevance or truth.
Related terms
Vector database
A store built for embeddings, so you can quickly find texts with a similar meaning.
Documents & retrievalSemantic search
Searching by meaning rather than by exact words, so 'delay' can find 'postponed'.
FoundationsTensor
A grid of numbers (a table, or a stack of tables). Everything inside a model is stored and calculated as tensors.
Where it is taught
How document-grounded AI answers workSource and further reading
Sentence Transformers · Semantic searchThe definition and example are written for this library. The link gives background, official documentation or legal authority as labelled. Legal scope depends on the jurisdiction and current text; product features depend on release and settings.