Usiing pdfplumber output for LLMs/RAG

[petermr]

I'm interested in using pdfplumber to create input for RAG/LLMs and search/analyse text [1]. Has anyone done anything in this area?

I have worked on heuristics to extract sections and paragraphs from pdfplumber which could be a useful input to a LLM.

[1] For example this is being done in the PyMUPDF project - PyMuPDF4LLM