{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/multilingual-extractive-reading-comprehension","title":"Multilingual Extractive Reading Comprehension by Runtime Machine Translation","arxiv_id":"1809.03275","date":"2018-09-10","proceeding":null,"authors":["Akari Asai","Akiko Eriguchi","Kazuma Hashimoto","Yoshimasa Tsuruoka"],"abstract":"Despite recent work in Reading Comprehension (RC), progress has been mostly\nlimited to English due to the lack of large-scale datasets in other languages.\nIn this work, we introduce the first RC system for languages without RC\ntraining data. Given a target language without RC training data and a pivot\nlanguage with RC training data (e.g. English), our method leverages existing RC\nresources in the pivot language by combining a competitive RC model in the\npivot language with an attentive Neural Machine Translation (NMT) model. We\nfirst translate the data from the target to the pivot language, and then obtain\nan answer using the RC model in the pivot language. Finally, we recover the\ncorresponding answer in the original language using soft-alignment attention\nscores from the NMT model. We create evaluation sets of RC data in two\nnon-English languages, namely Japanese and French, to evaluate our method.\nExperimental results on these datasets show that our method significantly\noutperforms a back-translation baseline of a state-of-the-art product-level\nmachine translation system.","url_abs":"http://arxiv.org/abs/1809.03275v2","url_pdf":"http://arxiv.org/pdf/1809.03275v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"multilingual-extractive-reading-comprehension","repo_url":"https://github.com/AkariAsai/extractive_rc_by_runtime_mt","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"nmt","task_name":"NMT"},{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.03275","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}