MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question Answering
AbstractProgress in cross-lingual modeling depends on challenging, realistic, and diverse evaluation sets. We introduce Multilingual Knowledge Questions and Answers (MKQA), an open- domain question answering evaluation set comprising 10k question-answer pairs aligned across 26 typolo...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
The MIT Press
2021-01-01
|
| Series: | Transactions of the Association for Computational Linguistics |
| Online Access: | https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00433/108607/MKQA-A-Linguistically-Diverse-Benchmark-for |
| _version_ | 1828527474713034752 |
|---|---|
| author | Shayne Longpre Yi Lu Joachim Daiber |
| author_facet | Shayne Longpre Yi Lu Joachim Daiber |
| author_sort | Shayne Longpre |
| collection | DOAJ |
| description |
AbstractProgress in cross-lingual modeling depends on challenging, realistic, and diverse evaluation sets. We introduce Multilingual Knowledge Questions and Answers (MKQA), an open- domain question answering evaluation set comprising 10k question-answer pairs aligned across 26 typologically diverse languages (260k question-answer pairs in total). Answers are based on heavily curated, language- independent data representation, making results comparable across languages and independent of language-specific passages. With 26 languages, this dataset supplies the widest range of languages to-date for evaluating question answering. We benchmark a variety of state- of-the-art methods and baselines for generative and extractive question answering, trained on Natural Questions, in zero shot and translation settings. Results indicate this dataset is challenging even in English, but especially in low-resource languages.1 |
| first_indexed | 2024-12-11T21:39:03Z |
| format | Article |
| id | doaj.art-1aa1d5711eac4e1cb516e3f093d5238f |
| institution | Directory Open Access Journal |
| issn | 2307-387X |
| language | English |
| last_indexed | 2024-12-11T21:39:03Z |
| publishDate | 2021-01-01 |
| publisher | The MIT Press |
| record_format | Article |
| series | Transactions of the Association for Computational Linguistics |
| spelling | doaj.art-1aa1d5711eac4e1cb516e3f093d5238f2022-12-22T00:49:53ZengThe MIT PressTransactions of the Association for Computational Linguistics2307-387X2021-01-0191389140610.1162/tacl_a_00433MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question AnsweringShayne Longpre0Yi Lu1Joachim Daiber2Apple Inc. slongpre@mit.eduApple Inc. ylu7@apple.comApple Inc. jodaiber@apple.com AbstractProgress in cross-lingual modeling depends on challenging, realistic, and diverse evaluation sets. We introduce Multilingual Knowledge Questions and Answers (MKQA), an open- domain question answering evaluation set comprising 10k question-answer pairs aligned across 26 typologically diverse languages (260k question-answer pairs in total). Answers are based on heavily curated, language- independent data representation, making results comparable across languages and independent of language-specific passages. With 26 languages, this dataset supplies the widest range of languages to-date for evaluating question answering. We benchmark a variety of state- of-the-art methods and baselines for generative and extractive question answering, trained on Natural Questions, in zero shot and translation settings. Results indicate this dataset is challenging even in English, but especially in low-resource languages.1https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00433/108607/MKQA-A-Linguistically-Diverse-Benchmark-for |
| spellingShingle | Shayne Longpre Yi Lu Joachim Daiber MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question Answering Transactions of the Association for Computational Linguistics |
| title | MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question Answering |
| title_full | MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question Answering |
| title_fullStr | MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question Answering |
| title_full_unstemmed | MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question Answering |
| title_short | MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question Answering |
| title_sort | mkqa a linguistically diverse benchmark for multilingual open domain question answering |
| url | https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00433/108607/MKQA-A-Linguistically-Diverse-Benchmark-for |
| work_keys_str_mv | AT shaynelongpre mkqaalinguisticallydiversebenchmarkformultilingualopendomainquestionanswering AT yilu mkqaalinguisticallydiversebenchmarkformultilingualopendomainquestionanswering AT joachimdaiber mkqaalinguisticallydiversebenchmarkformultilingualopendomainquestionanswering |