paper-with-me

홈 › Papers

Marian: Fast Neural Machine Translation in C++

2018-04-01 · ACL 2018 7 · Marcin Junczys-Dowmunt, Roman Grundkiewicz, Tomasz Dwojak, Hieu Hoang, Kenneth Heafield, Tom Neckermann, Frank Seide, Ulrich Germann, Alham Fikri Aji, Nikolay Bogoychev, André F. T. Martins, Alexandra Birch

We present Marian, an efficient and self-contained Neural Machine Translation framework with an integrated automatic differentiation engine based on dynamic computation graphs. Marian is written entirely in C++. We describe the design of the encoder-decoder framework and demonstrate that a research-friendly toolkit can achieve high training and translation speed.

📄 PDF Abstract BibTeX arXiv:1804.00344

Code (3)

marian-nmt/marian 공식 구현
emjotde/amunmt
pwc-1/Paper-9/tree/main/marian mindspore

Tasks

DecoderMachine TranslationTranslation

Similar Papers 제목 키워드 기반

PyMarian: Fast Neural Machine Translation and Evaluation in Python

2024-08-15 · Thamme Gowda, Roman Grundkiewicz, Elijah Rippeth, Matt Post 외

The deep learning language of choice these days is Python; measured by factors such as available libraries and technical support, it is hard to beat. At the same time, software written in lower-level programming language…

Machine TranslationNMTTranslation

MarianCG: a code generation transformer model inspired by machine translation

2022-11-22 · Journal of Engineering and Applied Science 2022 11 · Ahmed S. Soliman, Mayada M. Hadhoud, Samir I. Shaheen

The idea that computers can build their own programs is extremely significant, and many researchers are working on this challenge. Code generation is described as the process of generating executable code that can be run…

Code GenerationCode TranslationLanguage ModelingLanguage Modelling+3

Marian: Cost-effective High-Quality Neural Machine Translation in C++

2018-05-30 · WS 2018 7 · Marcin Junczys-Dowmunt, Kenneth Heafield, Hieu Hoang, Roman Grundkiewicz 외

This paper describes the submissions of the "Marian" team to the WNMT 2018 shared task. We investigate combinations of teacher-student training, low-precision matrix products, auto-tuning and other methods to optimize th…

CPUGPUMachine TranslationTranslation+1

Picard understanding Darmok: A Dataset and Model for Metaphor-Rich Translation in a Constructed Language

2021-07-16 · Peter Jansen, Jordan Boyd-Graber

Tamarian, a fictional language introduced in the Star Trek episode Darmok, communicates meaning through utterances of metaphorical references, such as "Darmok and Jalad at Tanagra" instead of "We should work together." T…

Language ModelingLanguage ModellingLarge Language ModelMachine Translation+1

TranslateLocally: Blazing-fast translation running on the local CPU

2021-09-21 · EMNLP (ACL) 2021 11 · Nikolay Bogoychev, Jelmer Van der Linde, Kenneth Heafield

Every day, millions of people sacrifice their privacy and browsing habits in exchange for online machine translation. Companies and governments with confidentiality requirements often ban online translation or pay a prem…

CPUMachine TranslationTranslation