{"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/augmenting-neural-machine-translation-with","title":"Augmenting Neural Machine Translation with Knowledge Graphs","arxiv_id":"1902.08816","date":"2019-02-23","proceeding":null,"authors":["Diego Moussallem","Mihael Arčan","Axel-Cyrille Ngonga Ngomo","Paul Buitelaar"],"abstract":"While neural networks have been used extensively to make substantial progress\nin the machine translation task, they are known for being heavily dependent on\nthe availability of large amounts of training data. Recent efforts have tried\nto alleviate the data sparsity problem by augmenting the training data using\ndifferent strategies, such as back-translation. Along with the data scarcity,\nthe out-of-vocabulary words, mostly entities and terminological expressions,\npose a difficult challenge to Neural Machine Translation systems. In this\npaper, we hypothesize that knowledge graphs enhance the semantic feature\nextraction of neural models, thus optimizing the translation of entities and\nterminological expressions in texts and consequently leading to a better\ntranslation quality. We hence investigate two different strategies for\nincorporating knowledge graphs into neural models without modifying the neural\nnetwork architectures. We also examine the effectiveness of our augmentation\nmethod to recurrent and non-recurrent (self-attentional) neural architectures.\nOur knowledge graph augmented neural translation model, dubbed KG-NMT, achieves\nsignificant and consistent improvements of +3 BLEU, METEOR and chrF3 on average\non the newstest datasets between 2014 and 2018 for WMT English-German\ntranslation task.","url_abs":"http://arxiv.org/abs/1902.08816v1","url_pdf":"http://arxiv.org/pdf/1902.08816v1.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":"augmenting-neural-machine-translation-with","repo_url":"https://github.com/dice-group/KG-NMT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"knowledge-graphs","task_name":"Knowledge Graphs"},{"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":null,"atlas_url":"https://app.syntology.ai/?focus=1902.08816","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}