language-models

Dense retrieval

taught

Finding relevant text by nearest-neighbor search over embedding vectors instead of keyword matching -- the mechanism every later RAG chapter builds on.

Taught in

Dense retrieval: why embeddings power search
Wiring Language Models

Requires

Nothing — this is foundational.

Required by

Multimodal embeddings and multimodal RAG required
The RAG pipeline shape required

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.