{"url":"/dataset/xnli","name":"XNLI","full_name":"Cross-lingual Natural Language Inference","description_markdown":"The **Cross-lingual Natural Language Inference** (**XNLI**) corpus is the extension of the Multi-Genre NLI (MultiNLI) corpus to 15 languages. The dataset was created by manually translating the validation and test sets of MultiNLI into each of those 15 languages. The English training set was machine translated for all languages. The dataset is composed of 122k train, 2490 validation and 5010 test examples.\r\n\r\nSource: [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894)\r\nImage Source: [https://github.com/facebookresearch/XNLI](https://github.com/facebookresearch/XNLI)","description_withheld":null,"homepage":"https://github.com/facebookresearch/XNLI","introduced_date":"2018-01-01","introduced_date_note":null,"introduced_by":{"paper":"/paper/xnli-evaluating-cross-lingual-sentence","title":"XNLI: Evaluating Cross-lingual Sentence Representations","first_author":"Alexis Conneau","url":null},"license":{"name":"Attribution-NonCommercial 4.0 International","url":"https://github.com/facebookresearch/XNLI/blob/master/LICENSE"},"modalities":[{"name":"Texts","url":"/datasets/modality/texts"}],"tasks":[{"name":"Natural Language Inference","url":"/task/natural-language-inference","datasets_with_task":"/datasets/task/natural-language-inference"},{"name":"Chinese Sentence Pair Classification","url":"/task/chinese-sentence-pair-classification","datasets_with_task":"/datasets/task/chinese-sentence-pair-classification"},{"name":"FLUE","url":"/task/flue","datasets_with_task":"/datasets/task/flue"},{"name":"Cross-Lingual Natural Language Inference","url":"/task/cross-lingual-natural-language-inference","datasets_with_task":"/datasets/task/cross-lingual-natural-language-inference"}],"languages":[{"name":"French","url":"/datasets/language/french"},{"name":"Spanish","url":"/datasets/language/spanish"},{"name":"German","url":"/datasets/language/german"},{"name":"Chinese","url":"/datasets/language/chinese"},{"name":"Multilingual","url":"/datasets/language/multilingual"},{"name":"Russian","url":"/datasets/language/russian"},{"name":"Arabic","url":"/datasets/language/arabic"},{"name":"Bulgarian","url":"/datasets/language/bulgarian"},{"name":"Hindi","url":"/datasets/language/hindi"},{"name":"Thai","url":"/datasets/language/thai"},{"name":"Turkish","url":"/datasets/language/turkish"},{"name":"Urdu","url":"/datasets/language/urdu"},{"name":"Vietnamese","url":"/datasets/language/vietnamese"},{"name":"Greek","url":"/datasets/language/greek"},{"name":"Swahili","url":"/datasets/language/swahili"}],"variants":["XNLI Zero-Shot English-to-French","XNLI French","XNLI Dev","XNLI Chinese","XNLI Chinese Dev","XNLI Zero-Shot English-to-German","XNLI Zero-Shot English-to-Spanish","XNLI"],"data_loaders":[{"repo":"https://github.com/huggingface/datasets","url":"https://huggingface.co/datasets/juletxara/xnli_mt","frameworks":["tf","pytorch","jax"]},{"repo":"https://github.com/huggingface/datasets","url":"https://huggingface.co/datasets/xnli","frameworks":["tf","pytorch","jax"]},{"repo":"https://github.com/huggingface/datasets","url":"https://huggingface.co/datasets/mteb/xnli","frameworks":["tf","pytorch","jax"]},{"repo":"https://github.com/huggingface/datasets","url":"https://huggingface.co/datasets/facebook/xnli","frameworks":["tf","pytorch","jax"]},{"repo":"https://github.com/huggingface/datasets","url":"https://huggingface.co/datasets/Mathoctopus/msvamp","frameworks":["tf","pytorch","jax"]},{"repo":"https://github.com/tensorflow/datasets","url":"https://www.tensorflow.org/datasets/catalog/xnli","frameworks":["tf","jax"]},{"repo":"https://github.com/facebookresearch/XNLI","url":"https://github.com/facebookresearch/XNLI","frameworks":[]}],"num_papers_in_archive":349,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/natural-language-inference-on-xnli-french","task":"Natural Language Inference","dataset_variant":"XNLI French","rows":6,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"CamemBERT (large)","paper":"/paper/camembert-a-tasty-french-language-model","metrics":{"Accuracy":"85.7"},"code_links":[{"title":"huggingface/transformers","url":"https://github.com/huggingface/transformers"},{"title":"Karthik-Bhaskar/Context-Based-Question-Answering","url":"https://github.com/Karthik-Bhaskar/Context-Based-Question-Answering"},{"title":"anaishoareau/french_preprocessing","url":"https://github.com/anaishoareau/french_preprocessing"},{"title":"bourrel/French-News-Clustering","url":"https://github.com/bourrel/French-News-Clustering"},{"title":"2024-MindSpore-1/Code2","url":"https://github.com/2024-MindSpore-1/Code2/tree/main/model-1/camembert"},{"title":"hbaflast/bert-sentiment-analysis-pytorch","url":"https://github.com/hbaflast/bert-sentiment-analysis-pytorch"},{"title":"hbaflast/bert-sentiment-analysis-tensorflow","url":"https://github.com/hbaflast/bert-sentiment-analysis-tensorflow"},{"title":"pwc-1/Paper-8","url":"https://github.com/pwc-1/Paper-8/tree/main/camembert"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cross-lingual-natural-language-inference-on-4","task":"Cross-Lingual Natural Language Inference","dataset_variant":"XNLI","rows":5,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"ByT5 XXL","paper":"/paper/byt5-towards-a-token-free-future-with-pre","metrics":{"Accuracy":"83.7"},"code_links":[{"title":"huggingface/transformers","url":"https://github.com/huggingface/transformers/tree/master/src/transformers/models/byt5"},{"title":"google-research/byt5","url":"https://github.com/google-research/byt5"},{"title":"ufal/multilexnorm2021","url":"https://github.com/ufal/multilexnorm2021"},{"title":"2024-MindSpore-1/Code2","url":"https://github.com/2024-MindSpore-1/Code2/tree/main/model-1/byt5"},{"title":"yoreG123/Paddle-ByT5","url":"https://github.com/yoreG123/Paddle-ByT5"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cross-lingual-natural-language-inference-on-1","task":"Cross-Lingual Natural Language Inference","dataset_variant":"XNLI Zero-Shot English-to-Spanish","rows":4,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"XLM-R R4F","paper":"/paper/better-fine-tuning-by-reducing","metrics":{"Accuracy":"85.2%"},"code_links":[{"title":"pytorch/fairseq","url":"https://github.com/pytorch/fairseq/tree/master/examples/rxf"},{"title":"cosmoquester/2021-dialogue-summary-competition","url":"https://github.com/cosmoquester/2021-dialogue-summary-competition"},{"title":"cliang1453/camero","url":"https://github.com/cliang1453/camero"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cross-lingual-natural-language-inference-on-3","task":"Cross-Lingual Natural Language Inference","dataset_variant":"XNLI Zero-Shot English-to-German","rows":4,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"XLM-R R4F","paper":"/paper/better-fine-tuning-by-reducing","metrics":{"Accuracy":"84.2%"},"code_links":[{"title":"pytorch/fairseq","url":"https://github.com/pytorch/fairseq/tree/master/examples/rxf"},{"title":"cosmoquester/2021-dialogue-summary-competition","url":"https://github.com/cosmoquester/2021-dialogue-summary-competition"},{"title":"cliang1453/camero","url":"https://github.com/cliang1453/camero"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cross-lingual-natural-language-inference-on","task":"Cross-Lingual Natural Language Inference","dataset_variant":"XNLI Zero-Shot English-to-French","rows":3,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"XLM-R R4F","paper":"/paper/better-fine-tuning-by-reducing","metrics":{"Accuracy":"84.7%"},"code_links":[{"title":"pytorch/fairseq","url":"https://github.com/pytorch/fairseq/tree/master/examples/rxf"},{"title":"cosmoquester/2021-dialogue-summary-competition","url":"https://github.com/cosmoquester/2021-dialogue-summary-competition"},{"title":"cliang1453/camero","url":"https://github.com/cliang1453/camero"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/natural-language-inference-on-xnli-chinese","task":"Natural Language Inference","dataset_variant":"XNLI Chinese Dev","rows":3,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"ERNIE 2.0 Large","paper":"/paper/ernie-20-a-continual-pre-training-framework","metrics":{"Accuracy":"82.6"},"code_links":[{"title":"PaddlePaddle/PaddleNLP","url":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/model_zoo/ernie-1.0"},{"title":"PaddlePaddle/ERNIE","url":"https://github.com/PaddlePaddle/ERNIE"},{"title":"DataScienceNigeria/ERNIE-2.0-from-Baidu-Inc.","url":"https://github.com/DataScienceNigeria/ERNIE-2.0-from-Baidu-Inc."}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/natural-language-inference-on-xnli-chinese-1","task":"Natural Language Inference","dataset_variant":"XNLI Chinese","rows":3,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"ERNIE 2.0 Large","paper":"/paper/ernie-20-a-continual-pre-training-framework","metrics":{"Accuracy":"81"},"code_links":[{"title":"PaddlePaddle/PaddleNLP","url":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/model_zoo/ernie-1.0"},{"title":"PaddlePaddle/ERNIE","url":"https://github.com/PaddlePaddle/ERNIE"},{"title":"DataScienceNigeria/ERNIE-2.0-from-Baidu-Inc.","url":"https://github.com/DataScienceNigeria/ERNIE-2.0-from-Baidu-Inc."}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/mgpt-few-shot-learners-go-multilingual","title":"mGPT: Few-Shot Learners Go Multilingual","date":"2022-04-15","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/byt5-towards-a-token-free-future-with-pre","title":"ByT5: Towards a token-free future with pre-trained byte-to-byte models","date":"2021-05-28","rows_on_this_dataset":2,"code_links":5,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":6,"samples_ran":0,"samples_unverified":6,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/rethinking-embedding-coupling-in-pre-trained-1","title":"Rethinking embedding coupling in pre-trained language models","date":"2020-10-24","rows_on_this_dataset":2,"code_links":4,"syntology":null},{"paper":"/paper/better-fine-tuning-by-reducing","title":"Better Fine-Tuning by Reducing Representational Collapse","date":"2020-08-06","rows_on_this_dataset":3,"code_links":3,"syntology":null},{"paper":"/paper/flaubert-unsupervised-language-model-pre","title":"FlauBERT: Unsupervised Language Model Pre-training for French","date":"2019-12-11","rows_on_this_dataset":2,"code_links":7,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":9,"samples_ran":1,"samples_unverified":8,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/camembert-a-tasty-french-language-model","title":"CamemBERT: a Tasty French Language Model","date":"2019-11-10","rows_on_this_dataset":2,"code_links":8,"syntology":null},{"paper":"/paper/ernie-20-a-continual-pre-training-framework","title":"ERNIE 2.0: A Continual Pre-training Framework for Language Understanding","date":"2019-07-29","rows_on_this_dataset":4,"code_links":3,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":1,"samples_ran":0,"samples_unverified":1,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/ernie-enhanced-representation-through","title":"ERNIE: Enhanced Representation through Knowledge Integration","date":"2019-04-19","rows_on_this_dataset":2,"code_links":19,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":7,"samples_ran":0,"samples_unverified":7,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/cross-lingual-language-model-pretraining","title":"Cross-lingual Language Model Pretraining","date":"2019-01-22","rows_on_this_dataset":1,"code_links":17,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":7,"samples_ran":1,"samples_unverified":6,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/bert-pre-training-of-deep-bidirectional","title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","date":"2018-10-11","rows_on_this_dataset":2,"code_links":534,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":659,"samples_ran":204,"samples_unverified":455,"pointer_only_for_licence":149,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/xnli-evaluating-cross-lingual-sentence","title":"XNLI: Evaluating Cross-lingual Sentence Representations","date":"2018-09-13","rows_on_this_dataset":1,"code_links":9,"syntology":null},{"paper":"/paper/supervised-learning-of-universal-sentence","title":"Supervised Learning of Universal Sentence Representations from Natural Language Inference Data","date":"2017-05-05","rows_on_this_dataset":6,"code_links":23,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":7,"samples_ran":6,"samples_unverified":1,"pointer_only_for_licence":7,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":7,"samples_harvested":696,"samples_ran":212,"samples_unverified":484,"pointer_only_for_licence":158,"papers_with_no_sample_that_ran":3,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}