100 datasets found

The dataset is used for training and testing the performance of multilingual language models.

Machine translation experiments on PADIC: A parallel Arabic dialect corpus

The dataset used in the paper is WMT'14 English-German, WAT'17 Japanese-English, and WMT'17 Chinese-English translation tasks.

OCR4MT is a benchmark for OCR systems on low-resource languages and scripts.

The dataset used for bilingual resynchronization task, which includes WMT14 English-French data and the small parallel sentence compression dataset.

The dataset used for bilingual synchronization task, which includes simulated interactive MT, translating with Translation Memory (TM) and TM cleaning.

The dataset used in the paper is a large corpus of text data, used for training and testing machine translation models.

A corpus of bilingual spontaneous written dialogues for machine translation.

The dataset used in the paper is not explicitly described, but it is mentioned that the authors used various datasets for machine translation tasks.

The dataset used in the paper is not explicitly described, but it is mentioned that the authors used the Moses toolkit to tokenize sentences and split words into subword units.

The dataset used in the paper is not explicitly described, but it is mentioned that the authors used four commonly-used benchmarks, including IT, Koran, Medical, and Law.

The IWSLT 2014 shared task dataset contains 152K, 156K, 141K and 172K training sentences for the de-en, zh-en, en-tr and en-es language pairs, respectively.

The dataset used in the paper is a subset of the OPUS-MT dataset, containing 1M randomly sampled examples from the OPUS-100 dataset.

Non-autoregressive machine translation dataset

The WMT14 En-De dataset contains 4.5M pairs of English and German sentences.

The paraphrased reference translations used for the experiments in the paper.

The paraphrased reference translations used for the experiments in the paper.

The dataset used for the experiments in the paper, containing English-German news translation task.

The IWSLT 2015 English-Vietnamese language data set, which has around 133k training sentence pairs.

The COMET dataset contains human-annotated scores for machine translation candidates.