Papers › Multilingual Denoising Pre-training for Neural Machine Translation

Multilingual Denoising Pre-training for Neural Machine Translation

22 Jan 2020arXiv:2001.08210archive 2025-07-28

Yinhan Liu, Jiatao Gu, Naman Goyal, Xi-An Li, Sergey Edunov, Marjan Ghazvininejad, Mike Lewis, Luke Zettlemoyer

This paper demonstrates that multilingual denoising pre-training produces significant performance gains across a wide variety of machine translation (MT) tasks. We present mBART -- a sequence-to-sequence denoising auto-encoder pre-trained on large-scale monolingual corpora in many languages using the BART objective. mBART is one of the first methods for pre-training a complete sequence-to-sequence model by denoising full texts in multiple languages, while previous approaches have focused only on the encoder, decoder, or reconstructing parts of the text. Pre-training a complete model allows it to be directly fine tuned for supervised (both sentence-level and document-level) and unsupervised machine translation, with no task-specific modifications. We demonstrate that adding mBART initialization produces performance gains in all but the highest-resource settings, including up to 12 BLEU points for low resource MT and over 5 BLEU points for many document-level and unsupervised models. We also show it also enables new types of transfer to language pairs with no bi-text or that were not in the pre-training corpus, and present extensive analysis of which factors contribute the most to effective pre-training.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

pytorch/fairseq officialpytorch report
avmb/marian-mbart mentioned on GitHub report
clmbrs/communication-translation mentioned on GitHubpytorch report
evtaktasheva/dependency_extraction mentioned on GitHubpytorch report
facebookresearch/GENRE mentioned on GitHubpytorch report
huggingface/transformers mentioned on GitHubpytorch report
hyunwoongko/asian-bart mentioned on GitHubpytorch report
pwc-1/Paper-9 mindspore report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

DecoderDenoisingMachine TranslationSentenceTranslationUnsupervised Machine Translation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AdamAttentionBARTBPEDense ConnectionsDropoutLayer NormalizationLinear LayerMulti-Head AttentionResidual ConnectionSoftmaxmBART

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections