BanglaSense: Bangla Text Summarization Using Generative AI with Large Language Model
Date
2023
Authors
Student ID
Research Supervisor
Editor
Journal Title
Volume
Issue
Journal Title
Journal ISSN
Volume Title
Abstract
Text summarization is used in various domains, such as generating concise summaries of news articles, documents including research papers, social media’s long-form posts, and conversations to quickly and easily understand a long text. On abstractive text summarization, due to the scarcity of datasets and computing resources for mid-level languages, recent efforts have concentrated mostly on high-resource languages like English. In this research, we propose a Bangla Text Summarization system Using Generative AI with a Large Language Model. We use in-context learning techniques and instruction fine-tuning to train the mT5 pre-trained multilingual Transformer model on the XL-Sum (Bangla) dataset from the Hugging Face library. Our fine-tuned model performed the best in terms of relevancy and correctness with a remarkable 45.41 rouge1, and 33.85 regulome Scores.
Description
Keywords
Citation
Department Name
Electrical and Computer Engineering
Publisher
North South University