dataset_infos.json

{"el-en": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"index": {"dtype": "int32", "id": null, "_type": "Value"}, "family_id": {"dtype": "int32", "id": null, "_type": "Value"}, "translation": {"languages": ["el", "en"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "el", "output": "en"}, "builder_name": "para_pat", "config_name": "el-en", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 24818840, "num_examples": 10855, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748818": {"num_bytes": 24894705, "checksum": "61122af21fbb51967dfa0e4f13c56c78259a8bed8d74a1be1bfd16f1f618f073"}}, "download_size": 24894705, "post_processing_size": null, "dataset_size": 24818840, "size_in_bytes": 49713545}, "cs-en": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"index": {"dtype": "int32", "id": null, "_type": "Value"}, "family_id": {"dtype": "int32", "id": null, "_type": "Value"}, "translation": {"languages": ["cs", "en"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "cs", "output": "en"}, "builder_name": "para_pat", "config_name": "cs-en", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 117555722, "num_examples": 78977, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748821": {"num_bytes": 118010340, "checksum": "732deb6a3994ae8839bd2e5b7ef6530a015f560fdc3d11c59f1d3e2ca43181da"}}, "download_size": 118010340, "post_processing_size": null, "dataset_size": 117555722, "size_in_bytes": 235566062}, "en-hu": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"index": {"dtype": "int32", "id": null, "_type": "Value"}, "family_id": {"dtype": "int32", "id": null, "_type": "Value"}, "translation": {"languages": ["en", "hu"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "en", "output": "hu"}, "builder_name": "para_pat", "config_name": "en-hu", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 80637157, "num_examples": 42629, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748827": {"num_bytes": 80893995, "checksum": "0090f3dc744ad6cb4e0f68c153a2ea4eb0b070e79be1fe826da194c49eba0a39"}}, "download_size": 80893995, "post_processing_size": null, "dataset_size": 80637157, "size_in_bytes": 161531152}, "en-ro": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"index": {"dtype": "int32", "id": null, "_type": "Value"}, "family_id": {"dtype": "int32", "id": null, "_type": "Value"}, "translation": {"languages": ["en", "ro"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "en", "output": "ro"}, "builder_name": "para_pat", "config_name": "en-ro", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 80290819, "num_examples": 48789, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748842": {"num_bytes": 80562562, "checksum": "4b749a0a778ce438334a354651e48e6e24fb132022d28ec35f2de564658b9528"}}, "download_size": 80562562, "post_processing_size": null, "dataset_size": 80290819, "size_in_bytes": 160853381}, "en-sk": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"index": {"dtype": "int32", "id": null, "_type": "Value"}, "family_id": {"dtype": "int32", "id": null, "_type": "Value"}, "translation": {"languages": ["en", "sk"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "en", "output": "sk"}, "builder_name": "para_pat", "config_name": "en-sk", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 31510348, "num_examples": 23410, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748848": {"num_bytes": 31707728, "checksum": "d180f2d49fd948cb56a181059db6ebb9ab56535279be32634b4443c7d433c213"}}, "download_size": 31707728, "post_processing_size": null, "dataset_size": 31510348, "size_in_bytes": 63218076}, "en-uk": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"index": {"dtype": "int32", "id": null, "_type": "Value"}, "family_id": {"dtype": "int32", "id": null, "_type": "Value"}, "translation": {"languages": ["en", "uk"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "en", "output": "uk"}, "builder_name": "para_pat", "config_name": "en-uk", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 136808871, "num_examples": 89226, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748851": {"num_bytes": 137391928, "checksum": "1b5d2b829d97ec7e1870a5017b5b6666ed4cfc677abae7be6104760edecdbb4c"}}, "download_size": 137391928, "post_processing_size": null, "dataset_size": 136808871, "size_in_bytes": 274200799}, "es-fr": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"index": {"dtype": "int32", "id": null, "_type": "Value"}, "family_id": {"dtype": "int32", "id": null, "_type": "Value"}, "translation": {"languages": ["es", "fr"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "es", "output": "fr"}, "builder_name": "para_pat", "config_name": "es-fr", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 53767035, "num_examples": 32553, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748857": {"num_bytes": 53989438, "checksum": "08d2464c27559204791e48782738c829b484d8cd7cbdcab6fcf78789e6cecf20"}}, "download_size": 53989438, "post_processing_size": null, "dataset_size": 53767035, "size_in_bytes": 107756473}, "fr-ru": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"index": {"dtype": "int32", "id": null, "_type": "Value"}, "family_id": {"dtype": "int32", "id": null, "_type": "Value"}, "translation": {"languages": ["fr", "ru"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "fr", "output": "ru"}, "builder_name": "para_pat", "config_name": "fr-ru", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 33915203, "num_examples": 10889, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748863": {"num_bytes": 33994490, "checksum": "0b478cd0eacab72c484a55fe7a4c93c234a2a7707237b7752815d8282f923fd3"}}, "download_size": 33994490, "post_processing_size": null, "dataset_size": 33915203, "size_in_bytes": 67909693}, "de-fr": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"translation": {"languages": ["de", "fr"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "de", "output": "fr"}, "builder_name": "para_pat", "config_name": "de-fr", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 655742822, "num_examples": 1167988, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748872": {"num_bytes": 204094654, "checksum": "6476bb81ce117e7df4d2cf0d7dd2680b7271ae05a74f9ef46be926377139d91a"}}, "download_size": 204094654, "post_processing_size": null, "dataset_size": 655742822, "size_in_bytes": 859837476}, "en-ja": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"translation": {"languages": ["en", "ja"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "en", "output": "ja"}, "builder_name": "para_pat", "config_name": "en-ja", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 3100002828, "num_examples": 6170339, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748626": {"num_bytes": 1093334863, "checksum": "4234bb5a07f09b2615c09ce9014a221a0c6859472b244d6b752f33ddf9d32c5c"}}, "download_size": 1093334863, "post_processing_size": null, "dataset_size": 3100002828, "size_in_bytes": 4193337691}, "en-es": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"translation": {"languages": ["en", "es"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "en", "output": "es"}, "builder_name": "para_pat", "config_name": "en-es", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 337690858, "num_examples": 649396, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748896": {"num_bytes": 105202237, "checksum": "dc9e7e6ca8bef81a3e560a9270565578ca99759ed0f714f720c516a105168849"}}, "download_size": 105202237, "post_processing_size": null, "dataset_size": 337690858, "size_in_bytes": 442893095}, "en-fr": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"translation": {"languages": ["en", "fr"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "en", "output": "fr"}, "builder_name": "para_pat", "config_name": "en-fr", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 6103179552, "num_examples": 12223525, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748944": {"num_bytes": 1846098331, "checksum": "527c34c760f9187e1630fce4bb33d830251b01d3b08147242d873b4d0493cbe9"}}, "download_size": 1846098331, "post_processing_size": null, "dataset_size": 6103179552, "size_in_bytes": 7949277883}, "de-en": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"translation": {"languages": ["de", "en"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "de", "output": "en"}, "builder_name": "para_pat", "config_name": "de-en", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 1059631418, "num_examples": 2165054, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23855657": {"num_bytes": 339299130, "checksum": "d165c3312c7218817d2c5afedfad68ebc8aee08f36ef4b798f7e620d055e4ea1"}}, "download_size": 339299130, "post_processing_size": null, "dataset_size": 1059631418, "size_in_bytes": 1398930548}, "en-ko": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"translation": {"languages": ["en", "ko"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "en", "output": "ko"}, "builder_name": "para_pat", "config_name": "en-ko", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 1466703472, "num_examples": 2324357, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748689": {"num_bytes": 475152089, "checksum": "4518f4c65391bf90fd153ade6eb23ca1393211852b9cc9d9e07c88d6a9ef2d04"}}, "download_size": 475152089, "post_processing_size": null, "dataset_size": 1466703472, "size_in_bytes": 1941855561}, "fr-ja": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"translation": {"languages": ["fr", "ja"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "fr", "output": "ja"}, "builder_name": "para_pat", "config_name": "fr-ja", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 211127021, "num_examples": 313422, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748866": {"num_bytes": 69038401, "checksum": "87376daa3e67806909122a2d102e8566eb25ea8523d7a24a95e15e1ec2dd4242"}}, "download_size": 69038401, "post_processing_size": null, "dataset_size": 211127021, "size_in_bytes": 280165422}, "en-zh": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"translation": {"languages": ["en", "zh"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "en", "output": "zh"}, "builder_name": "para_pat", "config_name": "en-zh", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 2297993338, "num_examples": 4897841, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748779": {"num_bytes": 899568201, "checksum": "cd8bed4124f7b6294e66aca5a31712f722be5bf2b28b908c054ad8d5c2e80fac"}}, "download_size": 899568201, "post_processing_size": null, "dataset_size": 2297993338, "size_in_bytes": 3197561539}, "en-ru": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"translation": {"languages": ["en", "ru"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "en", "output": "ru"}, "builder_name": "para_pat", "config_name": "en-ru", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 1974874480, "num_examples": 4296399, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23748704": {"num_bytes": 567240359, "checksum": "976656d19ecf009dbf9dbf733c69ba0730c21b30b0a0545e1b0f3f22a0de0504"}}, "download_size": 567240359, "post_processing_size": null, "dataset_size": 1974874480, "size_in_bytes": 2542114839}, "fr-ko": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"index": {"dtype": "int32", "id": null, "_type": "Value"}, "family_id": {"dtype": "int32", "id": null, "_type": "Value"}, "translation": {"languages": ["fr", "ko"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "fr", "output": "ko"}, "builder_name": "para_pat", "config_name": "fr-ko", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 222006786, "num_examples": 120607, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23855408": {"num_bytes": 64621605, "checksum": "6a33d0d1194afd005daacd82baf9305d71673d40b2db999fcd9f55c09175ca23"}}, "download_size": 64621605, "post_processing_size": null, "dataset_size": 222006786, "size_in_bytes": 286628391}, "ru-uk": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"index": {"dtype": "int32", "id": null, "_type": "Value"}, "family_id": {"dtype": "int32", "id": null, "_type": "Value"}, "translation": {"languages": ["ru", "uk"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "ru", "output": "uk"}, "builder_name": "para_pat", "config_name": "ru-uk", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 163442529, "num_examples": 85963, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23855465": {"num_bytes": 38709524, "checksum": "7c4cf896fac4df9f8a5b0b434b435be65136ecd017168833b1648ad035971890"}}, "download_size": 38709524, "post_processing_size": null, "dataset_size": 163442529, "size_in_bytes": 202152053}, "en-pt": {"description": "ParaPat: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\n\nThis dataset contains the developed parallel corpus from the open access Google\nPatents dataset in 74 language pairs, comprising more than 68 million sentences\nand 800 million tokens. Sentences were automatically aligned using the Hunalign algorithm\nfor the largest 22 language pairs, while the others were abstract (i.e. paragraph) aligned.\n\nWe demonstrate the capabilities of our corpus by training Neural Machine Translation\n(NMT) models for the main 9 language pairs, with a total of 18 models.\n", "citation": "@inproceedings{soares-etal-2020-parapat,\n    title = \"{P}ara{P}at: The Multi-Million Sentences Parallel Corpus of Patents Abstracts\",\n    author = \"Soares, Felipe  and\n      Stevenson, Mark  and\n      Bartolome, Diego  and\n      Zaretskaya, Anna\",\n    booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n    month = may,\n    year = \"2020\",\n    address = \"Marseille, France\",\n    publisher = \"European Language Resources Association\",\n    url = \"https://www.aclweb.org/anthology/2020.lrec-1.465\",\n    pages = \"3769--3774\",\n    language = \"English\",\n    ISBN = \"979-10-95546-34-4\",\n}\n", "homepage": "https://figshare.com/articles/ParaPat_The_Multi-Million_Sentences_Parallel_Corpus_of_Patents_Abstracts/12627632", "license": "CC BY 4.0", "features": {"index": {"dtype": "int32", "id": null, "_type": "Value"}, "family_id": {"dtype": "int32", "id": null, "_type": "Value"}, "translation": {"languages": ["en", "pt"], "id": null, "_type": "Translation"}}, "post_processed": null, "supervised_keys": {"input": "en", "output": "pt"}, "builder_name": "para_pat", "config_name": "en-pt", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 37372555, "num_examples": 23121, "dataset_name": "para_pat"}}, "download_checksums": {"https://ndownloader.figshare.com/files/23855441": {"num_bytes": 12781082, "checksum": "8cf90b1c45f20d4d9e0336cb77f1c200b932327f8ff9b82c9b9db329cf681c58"}}, "download_size": 12781082, "post_processing_size": null, "dataset_size": 37372555, "size_in_bytes": 50153637}}