{"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/fine-grained-human-evaluation-of-neural","title":"Fine-grained human evaluation of neural versus phrase-based machine translation","arxiv_id":"1706.04389","date":"2017-06-14","proceeding":null,"authors":["Filip Klubička","Antonio Toral","Víctor M. Sánchez-Cartagena"],"abstract":"We compare three approaches to statistical machine translation (pure\nphrase-based, factored phrase-based and neural) by performing a fine-grained\nmanual evaluation via error annotation of the systems' outputs. The error types\nin our annotation are compliant with the multidimensional quality metrics\n(MQM), and the annotation is performed by two annotators. Inter-annotator\nagreement is high for such a task, and results show that the best performing\nsystem (neural) reduces the errors produced by the worst system (phrase-based)\nby 54%.","url_abs":"http://arxiv.org/abs/1706.04389v1","url_pdf":"http://arxiv.org/pdf/1706.04389v1.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":"fine-grained-human-evaluation-of-neural","repo_url":"https://github.com/GreenParachute/mqm-eng-cro","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}