The Cross-lingual Natural Language Inference (XNLI) corpus is a crowd-sourced collection of 5,000 test and 2,500 dev pairs for the MultiNLI corpus. The pairs are annotated with textual entailment and translated into 14 languages: French, Spanish, German, Greek, Bulgarian, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, Hindi, Swahili and Urdu. This results in 112.5k annotated pairs. Each premise can be associated with the corresponding hypothesis in the 15 languages, summing up to more than 1.5M combinations. The corpus is made to evaluate how to perform inference in any language (including low-resources ones like Swahili or Urdu) when only English NLI data is available at training time. One solution is cross-lingual sentence encoding, for which XNLI is an evaluation benchmark.

Text excerpted from XNLI webpage on nyu.edu