diff --git "a/dataset_infos.json" "b/dataset_infos.json"
new file mode 100644--- /dev/null
+++ "b/dataset_infos.json"
@@ -0,0 +1 @@
+{"X-CSQA-en": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-en", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-zh": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-zh", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-de": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-de", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-es": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-es", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-fr": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-fr", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-it": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-it", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-jap": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-jap", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-nl": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-nl", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-pl": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-pl", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-pt": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-pt", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-ru": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-ru", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-ar": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-ar", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-vi": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-vi", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-hi": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-hi", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-sw": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-sw", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CSQA-ur": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CSQA-ur", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 721549, "num_examples": 1074, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 689431, "num_examples": 1000, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1410980, "size_in_bytes": 8930883}, "X-CODAH-en": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-en", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1764740, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 511779, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2276519, "size_in_bytes": 9796422}, "X-CODAH-zh": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-zh", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 832040, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 241465, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1073505, "size_in_bytes": 8593408}, "X-CODAH-de": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-de", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 2041351, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 590880, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2632231, "size_in_bytes": 10152134}, "X-CODAH-es": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-es", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1906430, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 548002, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2454432, "size_in_bytes": 9974335}, "X-CODAH-fr": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-fr", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 2024309, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 580389, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2604698, "size_in_bytes": 10124601}, "X-CODAH-it": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-it", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1956119, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 567652, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2523771, "size_in_bytes": 10043674}, "X-CODAH-jap": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-jap", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1173047, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 341716, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 1514763, "size_in_bytes": 9034666}, "X-CODAH-nl": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-nl", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1922420, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 552682, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2475102, "size_in_bytes": 9995005}, "X-CODAH-pl": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-pl", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1786230, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 516778, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2303008, "size_in_bytes": 9822911}, "X-CODAH-pt": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-pt", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1923583, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 552849, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2476432, "size_in_bytes": 9996335}, "X-CODAH-ru": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-ru", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1989775, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 568113, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2557888, "size_in_bytes": 10077791}, "X-CODAH-ar": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-ar", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1659286, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 479370, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2138656, "size_in_bytes": 9658559}, "X-CODAH-vi": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-vi", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1947493, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 560024, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2507517, "size_in_bytes": 10027420}, "X-CODAH-hi": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-hi", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 2365525, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 685704, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 3051229, "size_in_bytes": 10571132}, "X-CODAH-sw": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-sw", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1797165, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 526694, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2323859, "size_in_bytes": 9843762}, "X-CODAH-ur": {"description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n", "citation": "# X-CSR\n@inproceedings{lin-etal-2021-xcsr,\n    title = \"Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning\",\n    author = \"Lin, Bill Yuchen and Lee, Seyeon and Qiao, Xiaoyang and Ren, Xiang\",\n    booktitle = \"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL-IJCNLP 2021)\",\n    year = \"2021\",\n    note={to appear}\n}\n\n# CSQA\n@inproceedings{Talmor2019commonsenseqaaq,\n    address = {Minneapolis, Minnesota},\n    author = {Talmor, Alon  and Herzig, Jonathan  and Lourie, Nicholas and Berant, Jonathan},\n    booktitle = {Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)},\n    doi = {10.18653/v1/N19-1421},\n    pages = {4149--4158},\n    publisher = {Association for Computational Linguistics},\n    title = {CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge},\n    url = {https://www.aclweb.org/anthology/N19-1421},\n    year = {2019}\n}\n\n# CODAH\n@inproceedings{Chen2019CODAHAA,\n    address = {Minneapolis, USA},\n    author = {Chen, Michael  and D{'}Arcy, Mike  and Liu, Alisa  and Fernandez, Jared  and Downey, Doug},\n    booktitle = {Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for {NLP}},\n    doi = {10.18653/v1/W19-2008},\n    pages = {63--69},\n    publisher = {Association for Computational Linguistics},\n    title = {CODAH: An Adversarially-Authored Question Answering Dataset for Common Sense},\n    url = {https://www.aclweb.org/anthology/W19-2008},\n    year = {2019}\n}\n", "homepage": "https://inklab.usc.edu//XCSR/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "lang": {"dtype": "string", "id": null, "_type": "Value"}, "question_tag": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"feature": {"stem": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"label": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "xcsr", "config_name": "X-CODAH-ur", "version": {"version_str": "1.1.0", "description": "", "major": 1, "minor": 1, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 2056883, "num_examples": 1000, "dataset_name": "xcsr"}, "validation": {"name": "validation", "num_bytes": 595413, "num_examples": 300, "dataset_name": "xcsr"}}, "download_checksums": {"https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {"num_bytes": 7519903, "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"}}, "download_size": 7519903, "post_processing_size": null, "dataset_size": 2652296, "size_in_bytes": 10172199}}
\ No newline at end of file