This workshop will explore how Retrieval Augmented Generation (RAG) can help improve the use of LLM-generated responses. LLM models know the probability of the connection between words, but there is no knowledge about the meeting of the words. LLM does not know what is grounded to reality. The RAG attempts to ground the output to facts and reduce nonsense (hallucinations).

This is a hands-on workshop geared to the software developer.

Please bring your notebooks.

 Suggested readings:

1 : https://ai.meta.com/blog/retrieval-augmented-generation-streamlining-the-creation-of-intelligent-natural-language-processing-models/

2: https://research.ibm.com/blog/retrieval-augmented-generation-RAG

Tools for the workshop: Python and Google Colab.