A museum encodes every catalog card into a vector so a new visitor question can find the nearest cards. What is that encoding step?
Select an answer to reveal the explanation.
Short Explanation
Every catalog card becomes a vector so a visitor question can find the nearest cards. That encoding is sentence embeddings from an encoder. It is not decoder training, not a page-count timeout, and not a kernel the intern has to write.
Full Explanation
An encoder maps text to a vector space so later lookup can find nearest neighbors. That is not decoder training, not a page-count limit, and not a handwritten kernel.