wikiHow to rag

RAG practice repo

I want to build a RAG powered chatbot for wikiHow type questions

WikiHow articles need a TLDR section. My chatbot will attempt to prepare TLDR summaries (text-summarization) for these articles, alongwith answering questions (question-answering)

RAG techniques have become more sophisticated than a simple similarity search

I aim to implement hybrid search, rerankers [1, 2]

I will be methodical in choosing an appropriate embedding-model and vector database

vector db

redis [1, 2]

embedding model

msmarco-distilbert-base-v4 is widely used for embedding process for chatbots, semantic search.

ref

1. Using Redis for real-time RAG goes beyond a Vector Database