Hi all !
More of an idea exchange than a discussion, I'm looking into several techno's and Adobe PDF Extract looks very promising.
Has anyone used it for multiple PDF files to set up a RAG with an LLM ?
- I'm wondering about the size of the chunks (which vary greatly) and the interpretability of the tables. Did you need to convert them to Markdown (as they are exported in .csv by default) for the LLM to interpret them "correctly"?
AdChoices