106 datasets found

XNMT is a neural machine translation toolkit that focuses on modular code design, making it easy to swap in and out different parts of the model.

The MADAR dataset is a parallel corpus for low-resource languages.

The ArzEnSEG corpus is a morphologically annotated dataset for code-switched Egyptian Arabic-English.

The ArzEn parallel corpus consists of speech transcriptions gathered through informal interviews with bilingual Egyptian Arabic-English speakers, as well as their English...

The dataset for English-to-Chinese controlled machine translation.

The dataset for Chinese-to-English controlled machine translation.

The dataset for English controlled machine translation.

The dataset used in the paper is a collection of text for machine translation, using a single machine translation system for multiple language directions.

Findings of the 2023 conference on machine translation (WMT23)

Findings of the WMT 2023 shared task on automatic post-editing

A multilingual LLM for explaining and correcting translation errors

Europarl English Romanian dataset.

IWSLT Vietnamese→English and ACL Romanian→English datasets.

The dataset used in the paper is a French-English translation task.

The dataset used for Vietnamese diacritic restoration problem, consisting of 180,000 sentence pairs.

English-to-Turkish task of the IWSLT 2014 dataset

English-to-Portuguese task of the IWSLT 2014 dataset

English-to-German task of the IWSLT 2014 dataset

The dataset used in the paper is WMT 2010 and WMT 2012 datasets, which contain machine translation tasks.

The Zh-En multi-domain dataset consists of four balanced domains: news, patent, subtitles, and COVID-19.