# Arabic NLP

This is an Arabic restaurant review sentiment analysis model fine-tuned based on AraBERT, capable of classifying reviews as positive or negative.

This is an Arabic re-ranking model based on the BERT architecture, optimized for Arabic text relevance ranking tasks

An Arabic text generation model fine-tuned based on Llama-3-8B-Instruct, focusing on improving the coherence and accuracy of Arabic text.

JAIS-30B is a 30-billion-parameter bilingual (Arabic and English) large language model based on the GPT-3 architecture, utilizing ALiBi positional embedding technology, achieving state-of-the-art performance in Arabic tasks.

This is an Arabic question-answering system fine-tuned on the AraElectra model, specifically optimized for the Arabic SQuADv2.0 dataset, capable of handling both answerable and unanswerable questions.

Stanza is an efficient multilingual linguistic analysis toolkit that provides natural language processing capabilities from raw text to syntactic analysis and entity recognition.

Pre-trained Arabic BERT medium language model, trained on approximately 8.2 billion words of Arabic text resources

An efficient Arabic language model based on Funnel Transformer and ELECTRA objective, with low computational cost and superior performance

CAMeLBERT is a collection of BERT models optimized for Arabic language variants, with the CA version specifically pre-trained on Classical Arabic texts

CAMeLBERT is a collection of pre-trained models for Arabic NLP tasks. This model is the Modern Standard Arabic (MSA) variant, trained on 12.6 billion tokens

A sentiment analysis model for Algerian dialect fine-tuned based on AraBERT, achieving optimal performance on YouTube comment datasets

Pretrained model for Arabic NLP tasks, trained on a reduced-scale (1/16) Modern Standard Arabic (MSA) dataset