A multilingual model XLM- RoBERTa for the textual entailment of sequence pair - premise and hypothesis of 15 different languages using the MNLI and XNLI corpus.