Hasty Briefsbeta

GitHub - langchain-ai/rag-from-scratch

6 days ago
https://github.com/langchain-ai/rag-from-scratch
Copy Link
  • #LLM
  • #RAG
  • #Machine Learning
  • LLMs are trained on a fixed corpus, limiting access to private or recent information.
  • Fine-tuning can help but is often costly and not ideal for factual recall.
  • Retrieval augmented generation (RAG) expands an LLM's knowledge by using external documents.
  • RAG grounds LLM generation through in-context learning.
  • A video playlist and notebooks explain RAG basics, including indexing, retrieval, and generation.
AboutLogin