Bart Large Xsum

🚀 Bart model finetuned on xsum

This is a BART model fine - tuned on the xsum dataset. It is designed for text summarization tasks, with ROUGE scores greater than 22 on the xsum dataset.

🚀 Quick Start

• Documentation: Check out the official docs at https://huggingface.co/transformers/model_doc/bart.html.
• Fine - tuning: You can refer to the fine - tuning examples in examples/seq2seq/ (as of Aug 20, 2020).
• Variants: Search for distilbart for variants of this model.
• Paper: Read the related paper at https://arxiv.org/abs/1910.13461.

✨ Features

• Text Summarization: Specialized for summarization tasks, achieving good ROUGE scores on multiple datasets.
• Fine - tuned: Fine - tuned on the xsum dataset to improve performance.

📚 Documentation

• Model Details: The model is facebook/bart-large-xsum, fine - tuned on the xsum dataset for summarization tasks.
• Metrics: | Property | Details | |----------|---------| | Model Type | facebook/bart-large-xsum | | Training Data | cnn_dailymail, xsum, samsum |

📊 Performance Metrics

cnn_dailymail (test split)

xsum (test split)

samsum (train split)

samsum (test split)

📄 License

This project is licensed under the MIT license.