medium10 pts · first answer countsragchunkingembeddings

What late chunking does

"Late chunking" uses a long-context embedding model in a particular way. What is it?

A.Encode the whole document once, then pool token vectors per chunkB.Chunk the document only after the user asks a questionC.Embed each chunk separately and average all of themD.Wait until the index is full before chunking

Discussion 0

Join free to post in the discussion.

What late chunking does

Discussion 0

Related questions