{"url":"/dataset/mldoc","name":"MLDoc","full_name":"Multilingual Document Classification Corpus","description_markdown":"**Multilingual Document Classification Corpus** (**MLDoc**) is a cross-lingual document classification dataset covering English, German, French, Spanish, Italian, Russian, Japanese and Chinese. It is a subset of the Reuters Corpus Volume 2 selected according to the following design choices:\r\n\r\n* uniform class coverage: same number of examples for each class and language,\r\n* official train / development / test split: for each language a training data of different sizes (1K, 2K, 5K and 10K stories), a development (1K) and a test corpus (4K) are provided (with exception of Spanish and Russian with 9458 and 5216 training documents respectively.\r\n\r\nSource: [A Corpus for Multilingual Document Classification in Eight Languages](https://paperswithcode.com/paper/a-corpus-for-multilingual-document/)","description_withheld":null,"homepage":"https://github.com/facebookresearch/MLDoc","introduced_date":"2018-01-01","introduced_date_note":null,"introduced_by":{"paper":"/paper/a-corpus-for-multilingual-document","title":"A Corpus for Multilingual Document Classification in Eight Languages","first_author":"Holger Schwenk","url":null},"license":{"name":"Custom","url":"https://github.com/facebookresearch/MLDoc#benchmark-dataset-rcv1rcv2-indices"},"modalities":[{"name":"Texts","url":"/datasets/modality/texts"}],"tasks":[{"name":"Cross-Lingual Document Classification","url":"/task/cross-lingual-document-classification","datasets_with_task":"/datasets/task/cross-lingual-document-classification"},{"name":"Cross-Lingual Sentiment Classification","url":"/task/cross-lingual-sentiment-classification","datasets_with_task":"/datasets/task/cross-lingual-sentiment-classification"}],"languages":[{"name":"English","url":"/datasets/language/english"},{"name":"French","url":"/datasets/language/french"},{"name":"Spanish","url":"/datasets/language/spanish"},{"name":"German","url":"/datasets/language/german"},{"name":"Italian","url":"/datasets/language/italian"},{"name":"Chinese","url":"/datasets/language/chinese"},{"name":"Multilingual","url":"/datasets/language/multilingual"},{"name":"Japanese","url":"/datasets/language/japanese"},{"name":"Russian","url":"/datasets/language/russian"}],"variants":["MLDoc","MLDoc Zero-Shot German-to-French","MLDoc Zero-Shot English-to-Spanish","MLDoc Zero-Shot English-to-Russian","MLDoc Zero-Shot English-to-Japanese","MLDoc Zero-Shot English-to-Italian","MLDoc Zero-Shot English-to-German","MLDoc Zero-Shot English-to-French","MLDoc Zero-Shot English-to-Chinese"],"data_loaders":[{"repo":"https://github.com/facebookresearch/MLDoc","url":"https://github.com/facebookresearch/MLDoc","frameworks":[]}],"num_papers_in_archive":53,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/cross-lingual-document-classification-on-1","task":"Cross-Lingual Document Classification","dataset_variant":"MLDoc Zero-Shot English-to-Spanish","rows":6,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"XLMft UDA","paper":"/paper/bridging-the-domain-gap-in-cross-lingual","metrics":{"Accuracy":"96.8"},"code_links":[{"title":"laiguokun/xlu-data","url":"https://github.com/laiguokun/xlu-data"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cross-lingual-document-classification-on-2","task":"Cross-Lingual Document Classification","dataset_variant":"MLDoc Zero-Shot English-to-French","rows":6,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"XLMft UDA","paper":"/paper/bridging-the-domain-gap-in-cross-lingual","metrics":{"Accuracy":"96.05"},"code_links":[{"title":"laiguokun/xlu-data","url":"https://github.com/laiguokun/xlu-data"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cross-lingual-document-classification-on","task":"Cross-Lingual Document Classification","dataset_variant":"MLDoc Zero-Shot English-to-German","rows":5,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"XLMft UDA","paper":"/paper/bridging-the-domain-gap-in-cross-lingual","metrics":{"Accuracy":"96.95%"},"code_links":[{"title":"laiguokun/xlu-data","url":"https://github.com/laiguokun/xlu-data"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cross-lingual-document-classification-on-8","task":"Cross-Lingual Document Classification","dataset_variant":"MLDoc Zero-Shot English-to-Chinese","rows":5,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"XLMft UDA","paper":"/paper/bridging-the-domain-gap-in-cross-lingual","metrics":{"Accuracy":"93.32"},"code_links":[{"title":"laiguokun/xlu-data","url":"https://github.com/laiguokun/xlu-data"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cross-lingual-document-classification-on-9","task":"Cross-Lingual Document Classification","dataset_variant":"MLDoc Zero-Shot English-to-Russian","rows":5,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"XLMft UDA","paper":"/paper/bridging-the-domain-gap-in-cross-lingual","metrics":{"Accuracy":"89.7"},"code_links":[{"title":"laiguokun/xlu-data","url":"https://github.com/laiguokun/xlu-data"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cross-lingual-document-classification-on-10","task":"Cross-Lingual Document Classification","dataset_variant":"MLDoc Zero-Shot English-to-Italian","rows":4,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"MultiFiT, pseudo","paper":"/paper/multifit-efficient-multi-lingual-language","metrics":{"Accuracy":"76.02"},"code_links":[{"title":"piegu/language-models","url":"https://github.com/piegu/language-models"},{"title":"n-waves/multifit","url":"https://github.com/n-waves/multifit"},{"title":"TheophileBlard/french-sentiment-analysis-with-bert","url":"https://github.com/TheophileBlard/french-sentiment-analysis-with-bert"},{"title":"lukexyz/Language-Models","url":"https://github.com/lukexyz/Language-Models"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cross-lingual-document-classification-on-11","task":"Cross-Lingual Document Classification","dataset_variant":"MLDoc Zero-Shot English-to-Japanese","rows":3,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"MultiFiT, pseudo","paper":"/paper/multifit-efficient-multi-lingual-language","metrics":{"Accuracy":"69.57"},"code_links":[{"title":"piegu/language-models","url":"https://github.com/piegu/language-models"},{"title":"n-waves/multifit","url":"https://github.com/n-waves/multifit"},{"title":"TheophileBlard/french-sentiment-analysis-with-bert","url":"https://github.com/TheophileBlard/french-sentiment-analysis-with-bert"},{"title":"lukexyz/Language-Models","url":"https://github.com/lukexyz/Language-Models"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cross-lingual-document-classification-on-14","task":"Cross-Lingual Document Classification","dataset_variant":"MLDoc Zero-Shot German-to-French","rows":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"BiLSTM (Europarl)","paper":"/paper/a-corpus-for-multilingual-document","metrics":{"Accuracy":"75.45"},"code_links":[{"title":"n-waves/multifit","url":"https://github.com/n-waves/multifit"},{"title":"facebookresearch/MLDoc","url":"https://github.com/facebookresearch/MLDoc"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/bridging-the-domain-gap-in-cross-lingual","title":"Bridging the domain gap in cross-lingual document classification","date":"2019-09-16","rows_on_this_dataset":5,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":2,"samples_ran":1,"samples_unverified":1,"pointer_only_for_licence":2,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/multifit-efficient-multi-lingual-language","title":"MultiFiT: Efficient Multi-lingual Language Model Fine-tuning","date":"2019-09-10","rows_on_this_dataset":7,"code_links":4,"syntology":null},{"paper":"/paper/massively-multilingual-sentence-embeddings","title":"Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond","date":"2018-12-26","rows_on_this_dataset":7,"code_links":13,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":10,"samples_ran":4,"samples_unverified":6,"pointer_only_for_licence":4,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/a-corpus-for-multilingual-document","title":"A Corpus for Multilingual Document Classification in Eight Languages","date":"2018-05-24","rows_on_this_dataset":16,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":15,"samples_ran":1,"samples_unverified":14,"pointer_only_for_licence":1,"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":3,"samples_harvested":27,"samples_ran":6,"samples_unverified":21,"pointer_only_for_licence":7,"papers_with_no_sample_that_ran":0,"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."}