Hello everyone. My head is turning around (like the fools garden song from lemon tree) Because I'm seeking for the perfect RAG and I read about reranking and then about CRAG and then about Knowledge graphs, and then about text to SQL queries,... As far as I'm seeing, KG are the best and more accurate solution, and correct me if I'm wrong please it can't be mixed with Vector Databases, right. What about reranking. Please share your thoughts on your fundings and tests. I'll as soon as I do more tests. At the moment, Vector DB and reranking is what I'm using but I'm thinking on jumping towards KG.
what would you use when creating an assistant that automatically analyzes a customer message and its content... and extracts the information into a database? Would you use Vector DB or KG? the most important thing is truthful QA / Extraction
RAGs feeded with info like URL's, text files, PDFs etc. We already know that Llamaparse is the best for OCR PDF's, and of course text PDF's. Also we know (like anyone on this) that plain txt are the best. But we already know that if you are feeding the RAG with, let's say Company annual summarys you can't chunk it as you are losing info from an already a summary. You can't afford that loose. Therefore you have some options: make the chunk size bigger or use other tech like I said, SQL, Graph... That's why I'm seeking others experience
Interested question. I'm not yet extracting fields from questions or answers. Are you? I'd love you to tell me.the best way. I'm thinking of feeding a CRM with a Chatbot and your info would be really useful
i mean im definetely no expert so id love to hear some input from our experts @Logan M @Teemu But im using a simple Pydantic Parser and output_cls to ask a specific question (about the info that i want) that then gets parsed into a json output. This work okay... but i would hope for some better solutions as i need it to be 100% exact and i hate doing prompting
