# Multi-round optimization

An Arabic text summarization model fine-tuned based on AraT5v2, optimized for news and long-text summarization

This model is a fine-tuned Arabic-to-English translation model based on Helsinki-NLP/opus-mt-ar-en, achieving a BLEU score of 54.65 on the evaluation set.

Fine-tuned model based on BERT-base for text classification tasks, achieving an F1 score of 0.9313 on the evaluation set

A Sinhala speech recognition model fine-tuned based on openai/whisper-small, supporting conversion of Sinhala speech to text.

A Thai text summarization model fine-tuned based on the mT5 architecture, capable of generating concise summaries between 40-140 characters

A BERT-based model for detecting offensive language in Spanish, fine-tuned on a Spanish offensive language dataset.

BERT-small based model for classifying code and natural language with over 99% accuracy

A Hindi to English translation model fine-tuned based on the google/mt5-small pre-trained model

An Indonesian fake news classification model based on IndoBERT, fine-tuned on an unknown dataset with an accuracy of 80.59%

German text classification model fine-tuned based on bert-base-german-cased, performs excellently on the evaluation set