ParsiNLU: A Suite of Language Understanding Challenges for Persian. Khashabi, D., Cohan, A., Shakeri, S., Hosseini, P., Pezeshkpour, P., Alikhani, M., Aminnaseri, M., Bitaab, M., Brahman, F., Ghazarian, S., Gheini, M., Kabiri, A., Mahabagdi, R. K., Memarrast, O., Mosallanezhad, A., Noury, E., Raji, S., Rasooli, M. S., Sadeghi, S., Azer, E. S., Samghabadi, N. S., Shafaei, M., Sheybani, S., Tazarv, A., & Yaghoobzadeh, Y. Transactions of the Association for Computational Linguistics, 9:1147–1162, MIT Press, Cambridge, MA, 2021. Paper doi abstract bibtex Abstract Despite the progress made in recent years in addressing natural language understanding (NLU) challenges, the majority of this progress remains to be concentrated on resource-rich languages like English. This work focuses on Persian language, one of the widely spoken languages in the world, and yet there are few NLU datasets available for this language. The availability of high-quality evaluation datasets is a necessity for reliable assessment of the progress on different NLU tasks and domains. We introduce ParsiNLU, the first benchmark in Persian language that includes a range of language understanding tasks—reading comprehension, textual entailment, and so on. These datasets are collected in a multitude of ways, often involving manual annotations by native speakers. This results in over 14.5k new instances across 6 distinct NLU tasks. Additionally, we present the first results on state-of-the-art monolingual and multilingual pre-trained language models on this benchmark and compare them with human performance, which provides valuable insights into our ability to tackle natural language understanding challenges in Persian. We hope ParsiNLU fosters further research and advances in Persian language understanding.1
@article{khashabi-etal-2021-parsinlu,
title = "ParsiNLU: A Suite of Language Understanding Challenges for Persian",
author = "Khashabi, Daniel and
Cohan, Arman and
Shakeri, Siamak and
Hosseini, Pedram and
Pezeshkpour, Pouya and
Alikhani, Malihe and
Aminnaseri, Moin and
Bitaab, Marzieh and
Brahman, Faeze and
Ghazarian, Sarik and
Gheini, Mozhdeh and
Kabiri, Arman and
Mahabagdi, Rabeeh Karimi and
Memarrast, Omid and
Mosallanezhad, Ahmadreza and
Noury, Erfan and
Raji, Shahab and
Rasooli, Mohammad Sadegh and
Sadeghi, Sepideh and
Azer, Erfan Sadeqi and
Samghabadi, Niloofar Safi and
Shafaei, Mahsa and
Sheybani, Saber and
Tazarv, Ali and
Yaghoobzadeh, Yadollah",
journal = "Transactions of the Association for Computational Linguistics",
volume = "9",
year = "2021",
address = "Cambridge, MA",
publisher = "MIT Press",
url = "https://aclanthology.org/2021.tacl-1.68",
doi = "10.1162/tacl_a_00419",
pages = "1147--1162",
abstract = "Abstract Despite the progress made in recent years in addressing natural language understanding (NLU) challenges, the majority of this progress remains to be concentrated on resource-rich languages like English. This work focuses on Persian language, one of the widely spoken languages in the world, and yet there are few NLU datasets available for this language. The availability of high-quality evaluation datasets is a necessity for reliable assessment of the progress on different NLU tasks and domains. We introduce ParsiNLU, the first benchmark in Persian language that includes a range of language understanding tasks{---}reading comprehension, textual entailment, and so on. These datasets are collected in a multitude of ways, often involving manual annotations by native speakers. This results in over 14.5k new instances across 6 distinct NLU tasks. Additionally, we present the first results on state-of-the-art monolingual and multilingual pre-trained language models on this benchmark and compare them with human performance, which provides valuable insights into our ability to tackle natural language understanding challenges in Persian. We hope ParsiNLU fosters further research and advances in Persian language understanding.1",
}
Downloads: 0
{"_id":"wBNmB8fn9ufSf4rxE","bibbaseid":"khashabi-cohan-shakeri-hosseini-pezeshkpour-alikhani-aminnaseri-bitaab-etal-parsinluasuiteoflanguageunderstandingchallengesforpersian-2021","author_short":["Khashabi, D.","Cohan, A.","Shakeri, S.","Hosseini, P.","Pezeshkpour, P.","Alikhani, M.","Aminnaseri, M.","Bitaab, M.","Brahman, F.","Ghazarian, S.","Gheini, M.","Kabiri, A.","Mahabagdi, R. K.","Memarrast, O.","Mosallanezhad, A.","Noury, E.","Raji, S.","Rasooli, M. S.","Sadeghi, S.","Azer, E. S.","Samghabadi, N. S.","Shafaei, M.","Sheybani, S.","Tazarv, A.","Yaghoobzadeh, Y."],"bibdata":{"bibtype":"article","type":"article","title":"ParsiNLU: A Suite of Language Understanding Challenges for Persian","author":[{"propositions":[],"lastnames":["Khashabi"],"firstnames":["Daniel"],"suffixes":[]},{"propositions":[],"lastnames":["Cohan"],"firstnames":["Arman"],"suffixes":[]},{"propositions":[],"lastnames":["Shakeri"],"firstnames":["Siamak"],"suffixes":[]},{"propositions":[],"lastnames":["Hosseini"],"firstnames":["Pedram"],"suffixes":[]},{"propositions":[],"lastnames":["Pezeshkpour"],"firstnames":["Pouya"],"suffixes":[]},{"propositions":[],"lastnames":["Alikhani"],"firstnames":["Malihe"],"suffixes":[]},{"propositions":[],"lastnames":["Aminnaseri"],"firstnames":["Moin"],"suffixes":[]},{"propositions":[],"lastnames":["Bitaab"],"firstnames":["Marzieh"],"suffixes":[]},{"propositions":[],"lastnames":["Brahman"],"firstnames":["Faeze"],"suffixes":[]},{"propositions":[],"lastnames":["Ghazarian"],"firstnames":["Sarik"],"suffixes":[]},{"propositions":[],"lastnames":["Gheini"],"firstnames":["Mozhdeh"],"suffixes":[]},{"propositions":[],"lastnames":["Kabiri"],"firstnames":["Arman"],"suffixes":[]},{"propositions":[],"lastnames":["Mahabagdi"],"firstnames":["Rabeeh","Karimi"],"suffixes":[]},{"propositions":[],"lastnames":["Memarrast"],"firstnames":["Omid"],"suffixes":[]},{"propositions":[],"lastnames":["Mosallanezhad"],"firstnames":["Ahmadreza"],"suffixes":[]},{"propositions":[],"lastnames":["Noury"],"firstnames":["Erfan"],"suffixes":[]},{"propositions":[],"lastnames":["Raji"],"firstnames":["Shahab"],"suffixes":[]},{"propositions":[],"lastnames":["Rasooli"],"firstnames":["Mohammad","Sadegh"],"suffixes":[]},{"propositions":[],"lastnames":["Sadeghi"],"firstnames":["Sepideh"],"suffixes":[]},{"propositions":[],"lastnames":["Azer"],"firstnames":["Erfan","Sadeqi"],"suffixes":[]},{"propositions":[],"lastnames":["Samghabadi"],"firstnames":["Niloofar","Safi"],"suffixes":[]},{"propositions":[],"lastnames":["Shafaei"],"firstnames":["Mahsa"],"suffixes":[]},{"propositions":[],"lastnames":["Sheybani"],"firstnames":["Saber"],"suffixes":[]},{"propositions":[],"lastnames":["Tazarv"],"firstnames":["Ali"],"suffixes":[]},{"propositions":[],"lastnames":["Yaghoobzadeh"],"firstnames":["Yadollah"],"suffixes":[]}],"journal":"Transactions of the Association for Computational Linguistics","volume":"9","year":"2021","address":"Cambridge, MA","publisher":"MIT Press","url":"https://aclanthology.org/2021.tacl-1.68","doi":"10.1162/tacl_a_00419","pages":"1147–1162","abstract":"Abstract Despite the progress made in recent years in addressing natural language understanding (NLU) challenges, the majority of this progress remains to be concentrated on resource-rich languages like English. This work focuses on Persian language, one of the widely spoken languages in the world, and yet there are few NLU datasets available for this language. The availability of high-quality evaluation datasets is a necessity for reliable assessment of the progress on different NLU tasks and domains. We introduce ParsiNLU, the first benchmark in Persian language that includes a range of language understanding tasks—reading comprehension, textual entailment, and so on. These datasets are collected in a multitude of ways, often involving manual annotations by native speakers. This results in over 14.5k new instances across 6 distinct NLU tasks. Additionally, we present the first results on state-of-the-art monolingual and multilingual pre-trained language models on this benchmark and compare them with human performance, which provides valuable insights into our ability to tackle natural language understanding challenges in Persian. We hope ParsiNLU fosters further research and advances in Persian language understanding.1","bibtex":"@article{khashabi-etal-2021-parsinlu,\n title = \"ParsiNLU: A Suite of Language Understanding Challenges for Persian\",\n author = \"Khashabi, Daniel and\n Cohan, Arman and\n Shakeri, Siamak and\n Hosseini, Pedram and\n Pezeshkpour, Pouya and\n Alikhani, Malihe and\n Aminnaseri, Moin and\n Bitaab, Marzieh and\n Brahman, Faeze and\n Ghazarian, Sarik and\n Gheini, Mozhdeh and\n Kabiri, Arman and\n Mahabagdi, Rabeeh Karimi and\n Memarrast, Omid and\n Mosallanezhad, Ahmadreza and\n Noury, Erfan and\n Raji, Shahab and\n Rasooli, Mohammad Sadegh and\n Sadeghi, Sepideh and\n Azer, Erfan Sadeqi and\n Samghabadi, Niloofar Safi and\n Shafaei, Mahsa and\n Sheybani, Saber and\n Tazarv, Ali and\n Yaghoobzadeh, Yadollah\",\n journal = \"Transactions of the Association for Computational Linguistics\",\n volume = \"9\",\n year = \"2021\",\n address = \"Cambridge, MA\",\n publisher = \"MIT Press\",\n url = \"https://aclanthology.org/2021.tacl-1.68\",\n doi = \"10.1162/tacl_a_00419\",\n pages = \"1147--1162\",\n abstract = \"Abstract Despite the progress made in recent years in addressing natural language understanding (NLU) challenges, the majority of this progress remains to be concentrated on resource-rich languages like English. This work focuses on Persian language, one of the widely spoken languages in the world, and yet there are few NLU datasets available for this language. The availability of high-quality evaluation datasets is a necessity for reliable assessment of the progress on different NLU tasks and domains. We introduce ParsiNLU, the first benchmark in Persian language that includes a range of language understanding tasks{---}reading comprehension, textual entailment, and so on. These datasets are collected in a multitude of ways, often involving manual annotations by native speakers. This results in over 14.5k new instances across 6 distinct NLU tasks. Additionally, we present the first results on state-of-the-art monolingual and multilingual pre-trained language models on this benchmark and compare them with human performance, which provides valuable insights into our ability to tackle natural language understanding challenges in Persian. We hope ParsiNLU fosters further research and advances in Persian language understanding.1\",\n}\n\n","author_short":["Khashabi, D.","Cohan, A.","Shakeri, S.","Hosseini, P.","Pezeshkpour, P.","Alikhani, M.","Aminnaseri, M.","Bitaab, M.","Brahman, F.","Ghazarian, S.","Gheini, M.","Kabiri, A.","Mahabagdi, R. K.","Memarrast, O.","Mosallanezhad, A.","Noury, E.","Raji, S.","Rasooli, M. S.","Sadeghi, S.","Azer, E. S.","Samghabadi, N. S.","Shafaei, M.","Sheybani, S.","Tazarv, A.","Yaghoobzadeh, Y."],"bibbaseid":"khashabi-cohan-shakeri-hosseini-pezeshkpour-alikhani-aminnaseri-bitaab-etal-parsinluasuiteoflanguageunderstandingchallengesforpersian-2021","role":"author","urls":{"Paper":"https://aclanthology.org/2021.tacl-1.68"},"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://bibbase.org/f/vneMZcSKXZt6sz8KD/gheini.bib","dataSources":["e4Q8Z3dm3WgDhb6Dj"],"keywords":[],"search_terms":["parsinlu","suite","language","understanding","challenges","persian","khashabi","cohan","shakeri","hosseini","pezeshkpour","alikhani","aminnaseri","bitaab","brahman","ghazarian","gheini","kabiri","mahabagdi","memarrast","mosallanezhad","noury","raji","rasooli","sadeghi","azer","samghabadi","shafaei","sheybani","tazarv","yaghoobzadeh"],"title":"ParsiNLU: A Suite of Language Understanding Challenges for Persian","year":2021}