Summary of Arabic-nougat: Fine-tuning Vision Transformers For Arabic Ocr and Markdown Extraction, by Mohamed Rashad
Arabic-Nougat: Fine-Tuning Vision Transformers for Arabic OCR and Markdown Extraction
by Mohamed Rashad
First submitted to arxiv on: 19 Nov 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Arabic-Nougat is a suite of OCR models for converting Arabic book pages into structured Markdown text. Based on Meta’s Nougat architecture, the suite includes three specialized models: arabic-small-nougat, arabic-base-nougat, and arabic-large-nougat. These models are fine-tuned on a synthetic dataset, arabic-img2md, comprising 13.7k pairs of Arabic book pages and their Markdown representations. The Aranizer-PBE-86k tokenizer is designed for efficient tokenization, and torch.bfloat16 precision with Flash Attention 2 is used for optimized training and inference. The models achieve state-of-the-art performance, with arabic-large-nougat delivering the highest Markdown Structure Accuracy and the lowest Character Error Rate. Arabic-Nougat also releases a large-scale dataset containing 1.1 billion Arabic tokens extracted from over 8,500 books using the best-performing model, providing a valuable resource for Arabic OCR research. All models, datasets, and code are open-sourced and available at this URL. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Arabic-Nougat is a special kind of computer program that can read and translate old Arabic books into plain text. It uses a combination of existing technologies to make the process work better. The team who made it created three different versions of the program, each one good for a specific task. They also made a big dataset of words from over 8,500 books, which will help other researchers and scientists learn more about Arabic writing. |
Keywords
» Artificial intelligence » Attention » Inference » Precision » Tokenization » Tokenizer
