language-models
Sentence and document embeddings
taught
Extending a single word's embedding to a whole sentence or document -- pooling token embeddings or using a model trained specifically to produce one fixed-size vector per passage -- and why this is the representation retrieval systems actually search over.
Taught in
-
Word, sentence, and document embeddings
Wiring Language Models
Requires
-
word-embeddings
recommended
deep-learning ch.17 teaches the word-level embedding; this course's sentence/document embeddings are the same vector-space idea at a coarser granularity.
Required by
Nothing in the catalog needs this yet.
Local knowledge map
Local knowledge map
Click to expand
Solid arrows are required prerequisites; dashed are recommended or helpful. Lower rows are more foundational. Click any box to jump to that concept.