{"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/towards-two-dimensional-sequence-to-sequence","title":"Towards Two-Dimensional Sequence to Sequence Model in Neural Machine Translation","arxiv_id":"1810.03975","date":"2018-10-09","proceeding":"EMNLP 2018 10","authors":["Parnia Bahar","Christopher Brix","Hermann Ney"],"abstract":"This work investigates an alternative model for neural machine translation\n(NMT) and proposes a novel architecture, where we employ a multi-dimensional\nlong short-term memory (MDLSTM) for translation modeling. In the\nstate-of-the-art methods, source and target sentences are treated as\none-dimensional sequences over time, while we view translation as a\ntwo-dimensional (2D) mapping using an MDLSTM layer to define the correspondence\nbetween source and target words. We extend beyond the current sequence to\nsequence backbone NMT models to a 2D structure in which the source and target\nsentences are aligned with each other in a 2D grid. Our proposed topology shows\nconsistent improvements over attention-based sequence to sequence model on two\nWMT 2017 tasks, German$\\leftrightarrow$English.","url_abs":"http://arxiv.org/abs/1810.03975v1","url_pdf":"http://arxiv.org/pdf/1810.03975v1.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":"towards-two-dimensional-sequence-to-sequence","repo_url":"https://github.com/FlorianPfisterer/2D-LSTM-Seq2Seq","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"nmt","task_name":"NMT"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"two","task_name":"Vocal Bursts Valence Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1810.03975","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}