One big area for improvement is document parsing. The current implementation has slow OCR and still pretty low accuracy when it comes to data in tables. We should explore alternatives to see if there are any improvements that can be made.
Here are some options:
One big area for improvement is document parsing. The current implementation has slow OCR and still pretty low accuracy when it comes to data in tables. We should explore alternatives to see if there are any improvements that can be made.
Here are some options: