Transactions of the Association for Computational Linguistics · 2021 · 113 citations · 35 references
ComputingComputer ScienceComputational LinguisticsCorpus LinguisticsNatural Language ProcessingLanguage ModelingMachine TranslationMultilingual Natural Language ProcessingMultilingual ProcessingQuestion AnsweringCross-lingual Language ModelingCross-lingual Representation LearningLow-resource Machine TranslationMultilingual Question AnsweringOpen-domain Question AnsweringQuestion Answering CorporaQuestion Answering SystemsQuestion Answering Systems Evaluation
Progress in cross‑lingual modeling relies on challenging, realistic, and diverse evaluation sets. We introduce MKQA, an open‑domain QA evaluation set with 10 k question‑answer pairs aligned across 26 typologically diverse languages (260 k pairs total). MKQA’s answers are based on a heavily curated, language‑independent representation, and the authors benchmark state‑of‑the‑art generative and extractive QA models trained on Natural Questions in zero‑shot and translation settings. MKQA covers the widest range of languages to date for QA evaluation, and results show it is challenging even in English, especially for low‑resource languages.
Abstract Progress 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
35
Ashish Vaswani, Noam Shazeer, Niki Parmar et al. · 2025 · 6.6K citations · Full text
SQuAD: 100,000+ Questions for Machine Comprehension of Text
Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev et al. · 2016 · 6.3K citations · Full text
Cross-lingual Language Model Pretraining
Guillaume Lample, Alexis Conneau · arXiv (Cornell University) · 2019 · 1.6K citations · Full text