paper-with-me

홈 › Papers

Understanding Back-Translation at Scale

2018-08-28 · EMNLP 2018 10 · Sergey Edunov, Myle Ott, Michael Auli, David Grangier

An effective method to improve neural machine translation with monolingual data is to augment the parallel training corpus with back-translations of target language sentences. This work broadens the understanding of back-translation and investigates a number of methods to generate synthetic source sentences. We find that in all but resource poor settings back-translations obtained via sampling or noised beam outputs are most effective. Our analysis shows that sampling or noisy synthetic data gives a much stronger training signal than data generated by beam or greedy search. We also compare how synthetic data compares to genuine bitext and study various domain effects. Finally, we scale to hundreds of millions of monolingual sentences and achieve a new state of the art of 35 BLEU on the WMT'14 English-German test set.

📄 PDF Abstract BibTeX arXiv:1808.09381

Code (3)

pytorch/fairseq 공식 구현 pytorch
facebookresearch/fairseq pytorch
valentinmace/noisy-text

Tasks

Machine TranslationTranslation

Similar Papers 제목 키워드 기반

A Reinforcement Learning-driven Translation Model for Search-Oriented Conversational Systems

2018-08-29 · WS 2018 10 · Wafa Aissa, Laure Soulier, Ludovic Denoyer

Search-oriented conversational systems rely on information needs expressed in natural language (NL). We focus here on the understanding of NL expressions for building keyword-based queries. We propose a reinforcement-lea…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)Translation

LLM-based Translation Inference with Iterative Bilingual Understanding

2024-10-16 · Andong Chen, Kehai Chen, Yang Xiang, Xuefeng Bai 외

The remarkable understanding and generation capabilities of large language models (LLMs) have greatly improved translation performance. However, incorrect understanding of the sentence to be translated can degrade transl…

SentenceTranslation

Steering AI-Driven Personalization of Scientific Text for General Audiences

2024-11-15 · Taewook Kim, Dhruv Agarwal, Jordan Ackerman, Manaswi Saha

Digital media platforms (e.g., social media, science blogs) offer opportunities to communicate scientific content to general audiences at scale. However, these audiences vary in their scientific expertise, literacy level…

Back-translation for Large-Scale Multilingual Machine Translation

2021-09-17 · WMT (EMNLP) 2021 11 · Baohao Liao, Shahram Khadivi, Sanjika Hewavitharana

This paper illustrates our approach to the shared task on large-scale multilingual machine translation in the sixth conference on machine translation (WMT-21). This work aims to build a single multilingual translation sy…

Machine TranslationTranslation

A Comparison between Pre-training and Large-scale Back-translation for Neural Machine Translation

2021-08-01 · Findings (ACL) 2021 8 · Dandan Huang, Kun Wang, Yue Zhang
Machine TranslationTranslation