{"id":1178,"date":"2022-12-27T09:39:18","date_gmt":"2022-12-27T08:39:18","guid":{"rendered":"https:\/\/www-staging.systransoft.com\/?post_type=paper&#038;p=1178"},"modified":"2023-11-16T18:22:47","modified_gmt":"2023-11-16T17:22:47","slug":"bilingual-synchronization-restoring-translational-relationships-with-editing-operations","status":"publish","type":"paper","link":"https:\/\/www-staging.systransoft.com\/jp\/resources\/papers-and-publications\/bilingual-synchronization-restoring-translational-relationships-with-editing-operations\/","title":{"rendered":"Bilingual Synchronization: Restoring Translational Relationships with Editing Operations"},"content":{"rendered":"<p>Machine Translation (MT) is usually viewed as a one-shot process that generates the target language equivalent of some source text from scratch. We consider here a more general setting which assumes an initial target sequence, that must be transformed into a valid translation of the source, thereby restoring parallelism between source and target. For this bilingual synchronization task, we consider several architectures (both autoregressive and non-autoregressive) and training regimes, and experiment with multiple practical settings such as simulated interactive MT, translating with Translation Memory (TM) and TM cleaning. Our results suggest that one single generic edit-based system, once fine-tuned, can compare with, or even outperform, dedicated systems specifically trained for these tasks.<\/p>","protected":false},"excerpt":{"rendered":"<p>Machine Translation (MT) is usually viewed as a one-shot process that generates the target language equivalent of some source text from scratch. We consider here a more general setting which assumes an initial target sequence, that must be transformed into a valid translation of the source, thereby restoring parallelism between source and target. For this bilingual synchronization task, we consider several architectures (both autoregressive and non-autoregressive) and training regimes, and experiment with multiple practical settings such as simulated interactive MT, translating with Translation Memory (TM) and TM cleaning. Our results suggest that one single generic edit-based system, once fine-tuned, can compare with, or even outperform, dedicated systems specifically trained for these tasks.<\/p>","protected":false},"featured_media":0,"template":"","class_list":["post-1178","paper","type-paper","status-publish","hentry"],"acf":[],"_links":{"self":[{"href":"https:\/\/www-staging.systransoft.com\/jp\/wp-json\/wp\/v2\/paper\/1178","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www-staging.systransoft.com\/jp\/wp-json\/wp\/v2\/paper"}],"about":[{"href":"https:\/\/www-staging.systransoft.com\/jp\/wp-json\/wp\/v2\/types\/paper"}],"wp:attachment":[{"href":"https:\/\/www-staging.systransoft.com\/jp\/wp-json\/wp\/v2\/media?parent=1178"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}