{"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/learning-to-remember-translation-history-with","title":"Learning to Remember Translation History with a Continuous Cache","arxiv_id":"1711.09367","date":"2017-11-26","proceeding":"TACL 2018 1","authors":["Zhaopeng Tu","Yang Liu","Shuming Shi","Tong Zhang"],"abstract":"Existing neural machine translation (NMT) models generally translate\nsentences in isolation, missing the opportunity to take advantage of\ndocument-level information. In this work, we propose to augment NMT models with\na very light-weight cache-like memory network, which stores recent hidden\nrepresentations as translation history. The probability distribution over\ngenerated words is updated online depending on the translation history\nretrieved from the memory, endowing NMT models with the capability to\ndynamically adapt over time. Experiments on multiple domains with different\ntopics and styles show the effectiveness of the proposed approach with\nnegligible impact on the computational cost.","url_abs":"http://arxiv.org/abs/1711.09367v1","url_pdf":"http://arxiv.org/pdf/1711.09367v1.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":"learning-to-remember-translation-history-with","repo_url":"https://github.com/longyuewangdcu/tvsub","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"nmt","task_name":"NMT"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1711.09367","atlas_url":"https://app.syntology.ai/?focus=1711.09367","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}