{"id":15883,"date":"2016-05-17T23:55:05","date_gmt":"2016-05-17T20:55:05","guid":{"rendered":"http:\/\/hgpu.org\/?p=15883"},"modified":"2016-05-17T23:55:05","modified_gmt":"2016-05-17T20:55:05","slug":"attention-based-nmt-models-as-feature-functions-in-phrase-based-smt","status":"publish","type":"post","link":"https:\/\/hgpu.org\/?p=15883","title":{"rendered":"Attention-based NMT Models as Feature Functions in Phrase-based SMT"},"content":{"rendered":"<p>This paper describes the AMU-UEDIN submissions to the WMT 2016 shared task on news translation. We explore methods of decode-time integration of attention-based neural translation models with phrase-based statistical machine translation. Efficient batch-algorithms for GPU-querying are proposed and implemented. For English-Russian, the phrase-based system cannot surpass state-of-the-art stand-alone neural models. For the Russian-English task, our submission achieves the top BLEU result, outperforming the best pure-neural system by 1.1 BLEU points and our own phrase-based baseline by 1.6 BLEU. In follow-up experiments we improve these results by additional 0.7 BLEU.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>This paper describes the AMU-UEDIN submissions to the WMT 2016 shared task on news translation. We explore methods of decode-time integration of attention-based neural translation models with phrase-based statistical machine translation. Efficient batch-algorithms for GPU-querying are proposed and implemented. For English-Russian, the phrase-based system cannot surpass state-of-the-art stand-alone neural models. For the Russian-English task, our [&hellip;]<\/p>\n","protected":false},"author":351,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[36,11,89,3],"tags":[1787,1782,14,1025,1815,20,1779],"class_list":["post-15883","post","type-post","status-publish","format-standard","hentry","category-algorithms","category-computer-science","category-nvidia-cuda","category-paper","tag-algorithms","tag-computer-science","tag-cuda","tag-machine-learning","tag-nlp","tag-nvidia","tag-nvidia-geforce-gtx-970"],"views":2213,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/15883","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/users\/351"}],"replies":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=15883"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/15883\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=15883"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=15883"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=15883"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}