2 ms·
An LLM with billions of parameters for extracting text from a PDF (which isn't even a rasterized image) really does not "solve OCR".
by einpoklum 2y ago
An LLM with billions of parameters for extracting text from a PDF (which isn't even a rasterized image) really does not "solve OCR".