paper-with-me

Papers

Massively Multilingual Neural Machine Translation in the Wild: Findings and Challenges

2019-07-11 · Naveen Arivazhagan, Ankur Bapna, Orhan Firat, Dmitry Lepikhin, Melvin Johnson, Maxim Krikun, Mia Xu Chen, Yuan Cao, George Foster, Colin Cherry, Wolfgang Macherey, Zhifeng Chen, Yonghui Wu

We introduce our efforts towards building a universal neural machine translation (NMT) system capable of translating between any language pair. We set a milestone towards this goal by building a single massively multilingual NMT model handling 103 languages trained on over 25 billion examples. Our system demonstrates effective transfer learning ability, significantly improving translation quality of low-resource languages, while keeping high-resource language translation quality on-par with competitive bilingual baselines. We provide in-depth analysis of various aspects of model building that are crucial to achieving quality and practicality in universal NMT. While we prototype a high-quality universal translation system, our extensive empirical analysis exposes issues that need to be further addressed, and we suggest directions for future research.

📄 PDF Abstract BibTeX arXiv:1907.05019

Code (0)

등록된 구현이 없습니다.

Tasks

Machine TranslationNMTTransfer LearningTranslation

Similar Papers 제목 키워드 기반

Massively Multilingual Neural Machine Translation

2019-02-28 · NAACL 2019 6 · Roee Aharoni, Melvin Johnson, Orhan Firat

Multilingual neural machine translation (NMT) enables training a single model that supports translation from multiple source languages into multiple target languages. In this paper, we push the limits of multilingual NMT…

Machine TranslationNMTTranslation

Hallucinations in Large Multilingual Translation Models

2023-03-28 · Nuno M. Guerreiro, Duarte Alves, Jonas Waldendorf, Barry Haddow 외

Large-scale multilingual machine translation systems have demonstrated remarkable ability to translate directly between numerous languages, making them increasingly appealing for real-world applications. However, when de…

Language ModelingLanguage ModellingLarge Language ModelMachine Translation+1

Evaluating the Cross-Lingual Effectiveness of Massively Multilingual Neural Machine Translation

2019-09-01 · Aditya Siddhant, Melvin Johnson, Henry Tsai, Naveen Arivazhagan 외

The recently proposed massively multilingual neural machine translation (NMT) system has been shown to be capable of translating over 100 languages to and from English within a single model. Its improved translation perf…

Cross-Lingual TransferMachine TranslationNMTTransfer Learning+1

Simple, Scalable Adaptation for Neural Machine Translation

2019-09-18 · IJCNLP 2019 11 · Ankur Bapna, Naveen Arivazhagan, Orhan Firat

Fine-tuning pre-trained Neural Machine Translation (NMT) models is the dominant approach for adapting to new languages and domains. However, fine-tuning requires adapting and maintaining a separate model for each target …

Domain AdaptationMachine TranslationNMTTranslation

Improving Massively Multilingual Neural Machine Translation and Zero-Shot Translation

2020-04-24 · ACL 2020 6 · Biao Zhang, Philip Williams, Ivan Titov, Rico Sennrich

Massively multilingual models for neural machine translation (NMT) are theoretically attractive, but often underperform bilingual models and deliver poor zero-shot translations. In this paper, we explore ways to improve …

Machine TranslationNMTTranslation