Looking for Up-to-Date Multimodal RAG cookbook with citations and highlighting

Hi all, I'm new to coding and looking for a comprehensive multimodal RAG cookbook that doesn't lock me into specific frameworks. My key requirements:

Processing both PDFs and Excel files Precise citation capabilities (showing exactly where in documents information was retrieved) Highlighting or bounding box functionality to visually indicate sources

I tried finding Google's cookbook on this topic but hit a 404 error. Before I start piecing together solutions from various sources, I'm wondering if anyone has recommendations for tutorials or resources that cover these needs with minimal framework lock-in. Thanks