nllb-200-distilled-600M-dz-to-en Open-Source Translation Model - Efficiently Achieve Arabic-to-English Translation

Home

Nllb 200 Distilled 600M Dz To En

Developed by KarmaCST

This model is a fine-tuned Arabic (Dz) to English translation model based on the distilled version of NLLB-200

Machine Translation

Transformers

#Low-resource translation #Arabic dialect translation #Distilled model

Downloads 17

Release Time : 4/13/2023

Model Overview

Fine-tuned from Meta's distilled NLLB-200 model, specifically designed for Arabic dialect (Dz) to English machine translation tasks

Model Features

High-quality dialect translation

Translation capability specifically optimized for Arabic dialect (Dz)

Distilled model efficiency

Based on distillation technology, reducing computational resource requirements while maintaining performance

Multi-round fine-tuning optimization

After 5 rounds of fine-tuning, significant improvement in BLEU score

Model Capabilities

Text translation

Cross-language conversion

Low-resource language processing

Use Cases

Content localization

Dialect content translation

Translate Arabic dialect content into English

BLEU score 42.87

Cross-language communication

Real-time chat translation

Support real-time communication for Arabic dialect speakers

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.9294	1.0	1688	0.8364	39.0175	13.2637
0.7929	2.0	3376	0.7893	40.9994	13.303
0.7069	3.0	5064	0.7737	42.4125	13.292
0.6482	4.0	6752	0.7701	42.826	13.3287
0.6231	5.0	8440	0.7727	42.8708	13.335

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご

Nllb 200 Distilled 600M Dz To En

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 nllb-200-distilled-600M-dz-to-en

🚀 Quick Start

📚 Documentation

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

📄 License