Closed olumolu closed 1 month ago
Using ocr https://github.com/ocrmypdf/OCRmyPDF And from pdf to txt. Maybe directly build ocr https://github.com/tesseract-ocr/tesseract And solve maths problems with new mathstral or other good models.
In general most models can understand latex (a language made for math annotation), in case the user doesn't use that, llava can read text directly from images, I think that could be a solution for this problem
Using ocr https://github.com/ocrmypdf/OCRmyPDF And from pdf to txt. Maybe directly build ocr https://github.com/tesseract-ocr/tesseract And solve maths problems with new mathstral or other good models.