BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension

Mike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad, Abdelrahman Mohamed, Omer Levy, Veselin Stoyanov, Luke Zettlemoyer

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We present BART, a denoising autoencoder for pretraining sequence-to-sequence models. BART is trained by (1) corrupting text with an arbitrary noising function, and (2) learning a model to reconstruct the original text. It uses a standard Tranformer-based neural machine translation architecture which, despite its simplicity, can be seen as generalizing BERT (due to the bidirectional encoder), GPT (with the left-to-right decoder), and other recent pretraining schemes. We evaluate a number of noising approaches, finding the best performance by both randomly shuffling the order of sentences and using a novel in-filling scheme, where spans of text are replaced with a single mask token. BART is particularly effective when fine tuned for text generation but also works well for comprehension tasks. It matches the performance of RoBERTa on GLUE and SQuAD, and achieves new state-of-the-art results on a range of abstractive dialogue, question answering, and summarization tasks, with gains of up to 3.5 ROUGE. BART also provides a 1.1 BLEU increase over a back-translation system for machine translation, with only target language pretraining. We also replicate other pretraining schemes within the BART framework, to understand their effect on end-task performance.
Original languageEnglish
Title of host publicationProceedings of the 58th Annual Meeting of the Association for Computational Linguistics
EditorsDan Jurafsky, Joyce Chai, Natalie Schluter, Joel Tetreault
PublisherAssociation for Computational Linguistics
Pages7871-7880
Number of pages10
ISBN (Electronic)978-1-952148-25-5
ISBN (Print)9781713813712
DOIs
StatePublished - 1 Jul 2020
Externally publishedYes
Event58th annual meeting of the Association for Computational Linguistics, ACL 2020 - Virtual
Duration: 5 Jul 202010 Jul 2020
Conference number: 58

Conference

Conference58th annual meeting of the Association for Computational Linguistics, ACL 2020
Abbreviated titleACL 2020
Period5/07/2010/07/20

Fingerprint

Dive into the research topics of 'BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension'. Together they form a unique fingerprint.

Cite this