{"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/a-shared-attention-mechanism-for","title":"A Shared Attention Mechanism for Interpretation of Neural Automatic Post-Editing Systems","arxiv_id":"1807.00248","date":"2018-07-01","proceeding":"WS 2018 7","authors":["Inigo Jauregi Unanue","Ehsan Zare Borzeshi","Massimo Piccardi"],"abstract":"Automatic post-editing (APE) systems aim to correct the systematic errors\nmade by machine translators. In this paper, we propose a neural APE system that\nencodes the source (src) and machine translated (mt) sentences with two\nseparate encoders, but leverages a shared attention mechanism to better\nunderstand how the two inputs contribute to the generation of the post-edited\n(pe) sentences. Our empirical observations have showed that when the mt is\nincorrect, the attention shifts weight toward tokens in the src sentence to\nproperly edit the incorrect translation. The model has been trained and\nevaluated on the official data from the WMT16 and WMT17 APE IT domain\nEnglish-German shared tasks. Additionally, we have used the extra 500K\nartificial data provided by the shared task. Our system has been able to\nreproduce the accuracies of systems trained with the same data, while at the\nsame time providing better interpretability.","url_abs":"http://arxiv.org/abs/1807.00248v1","url_pdf":"http://arxiv.org/pdf/1807.00248v1.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":"a-shared-attention-mechanism-for","repo_url":"https://github.com/ijauregiCMCRC/Shared_Attention_for_APE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"automatic-post-editing","task_name":"Automatic Post-Editing"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}