Study
LLMs · Embeddings
Sentence embeddings for semantic search: mean-pooling a small encoder (all-MiniLM-L6-v2), cosine similarity, and finding the most relevant sentence for a query without any shared keywords.
- An embedding model maps a whole sentence to one vector, so similar meanings land close together even with different words.
- This is the retrieval half of RAG, which the LangChain phase builds on.