alvations/xnli-15way

XNLI consists of 10k English sentences translated into 14 languages:

ar: Arabic
bg: Bulgarian
de: German
el: Greek
es: Spanish
fr: French
hi: Hindi
ru: Russian
sw: Swahili
th: Thai
tr: Turkish
ur: Urdu
vi: Vietnamese
zh: Chinese (Simplified)

The XNLI 15-way parallel corpus can be used for Machine Translation as evaluation sets, in particular for low-resource languages such as Swahili or Urdu.

We provide two files: xnli.15way.orig.tsv and xnli.15way.tok.tsv containing respectively the original and the tokenized version of the corpus.
The files consist of 15 tab-separated columns, each corresponding to one language as indicated by the header.

Please consider citing the following paper if using this dataset:

@InProceedings{conneau2018xnli,
author = "Conneau, Alexis
and Rinott, Ruty
and Lample, Guillaume
and Williams, Adina
and Bowman, Samuel R.
and Schwenk, Holger
and Stoyanov, Veselin",
title = "XNLI: Evaluating Cross-lingual Sentence Representations",
booktitle = "Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing",
year = "2018",
publisher = "Association for Computational Linguistics",
location = "Brussels, Belgium",
}

原始信息汇总

数据集概述

数据集名称

XNLI

数据集内容

• 语言种类：包含15种语言，具体包括：

  • Arabic (ar)
  • Bulgarian (bg)
  • German (de)
  • Greek (el)
  • Spanish (es)
  • French (fr)
  • Hindi (hi)
  • Russian (ru)
  • Swahili (sw)
  • Thai (th)
  • Turkish (tr)
  • Urdu (ur)
  • Vietnamese (vi)
  • Chinese (Simplified) (zh)

• 数据用途：适用于机器翻译评估，特别是针对低资源语言如Swahili或Urdu。

数据集文件

• 文件1：xnli.15way.orig.tsv - 包含原始语料。
• 文件2：xnli.15way.tok.tsv - 包含已分词的语料。
• 文件格式：15列，每列对应一种语言，以制表符分隔。

引用信息

• 论文：Conneau, Alexis et al. "XNLI: Evaluating Cross-lingual Sentence Representations". Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics, 2018.
• 出版信息：Brussels, Belgium.