{"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/has-machine-translation-achieved-human-parity","title":"Has Machine Translation Achieved Human Parity? A Case for Document-level Evaluation","arxiv_id":"1808.07048","date":"2018-08-21","proceeding":"EMNLP 2018 10","authors":["Samuel Läubli","Rico Sennrich","Martin Volk"],"abstract":"Recent research suggests that neural machine translation achieves parity with\nprofessional human translation on the WMT Chinese--English news translation\ntask. We empirically test this claim with alternative evaluation protocols,\ncontrasting the evaluation of single sentences and entire documents. In a\npairwise ranking experiment, human raters assessing adequacy and fluency show a\nstronger preference for human over machine translation when evaluating\ndocuments as compared to isolated sentences. Our findings emphasise the need to\nshift towards document-level evaluation as machine translation improves to the\ndegree that errors which are hard or impossible to spot at the sentence-level\nbecome decisive in discriminating quality of different translation outputs.","url_abs":"http://arxiv.org/abs/1808.07048v1","url_pdf":"http://arxiv.org/pdf/1808.07048v1.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":"has-machine-translation-achieved-human-parity","repo_url":"https://github.com/laeubli/parity","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.07048","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}