M

Mbert2mbert Arabic Text Summarization

Developed by malmarjeh
This is an Arabic text summarization model based on the BERT2BERT architecture, initialized with mBERT weights and fine-tuned on a dataset of 84,764 paragraph-summary pairs.
Downloads 211
Release Time : 6/3/2022

Model Overview

This model is specifically designed for abstract summarization, news title generation, and paraphrasing tasks of Arabic text.

Model Features

Initialized based on mBERT
Initialized with the weights of Multilingual BERT (mBERT), which improves the model's understanding of Arabic.
Large-scale fine-tuning data
Fine-tuned on a dataset of 84,764 Arabic paragraph-summary pairs.
Multifunctional application
Not only supports text summarization but also can be used for news title generation and text paraphrasing.

Model Capabilities

Arabic text summarization
Arabic news title generation
Arabic text paraphrasing

Use Cases

News media
News summary generation
Automatically generate a short summary of a news article.
As shown in the example, it can compress a long news paragraph into a concise summary.
Content creation
Text restatement
Paraphrase and rewrite Arabic text.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase