BART: Denoising Sequence-to-Sequence Pre-training for NLG & Translation (Explained)

BART is a powerful model that can be used for many different text generation tasks, including summarization, machine translation, and abstract question answering. It could also be used for text classification and token classification. This video explains the architecture of BART and how it leverages 6 different pre-training objectives to achieve excellence.

BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension https://arxiv.org/abs/1910.13461

Code (Facebook) https://github.com/pytorch/fairseq/tree/main/examples/bart

Code (Hugginface) https://huggingface.co/transformers/model_doc/bart.html

Connect Linkedin https://www.linkedin.com/in/xue-yong-fu-955723a6/ Twitter https://twitter.com/home email edwindeeplearning@gmail.com ... https://www.youtube.com/watch?v=MxNnl_gHV1Y