File size: 1,172 Bytes
28855be
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cc92bbf
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
28855be
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
---
language:
- ar
- bg
- de
- el
- es
- fr
- hi
- ru
- sw
- th
- tr
- ur
- vi
- zh
---
XNLI consists of 10k English sentences translated into 14 languages: 


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",
}