Beyond Prefix-Based Interactive Translation Prediction

Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning (CoNLL), pages 198–207, Berlin, Germany, August 7-12, 2016. c⃝2016 Association for Computational Linguistics Beyond Prefix-Based Interactive Translation Prediction Jes´us Gonz´alez-Rubio Daniel Ortiz-Mart´ınez Webinterpret Inc. {jesus.g,daniel.o}@webinterpret.com Jos´e Miguel Bened´ı Francisco Casacuberta PRHLT Research Center Univ. Polit`ecnica de Val`encia {jbenedi,fcn}@prhlt.upv.es Abstract Current automatic machine translation systems require heavy human proof- reading to produce high-quality transla- tions. We present a new interactive ma- chine translation approach aimed at pro- viding a natural collaboration between hu- mans and translation systems. As such, we grant the user complete freedom to vali- date and correct any part of the translations suggested by the system. Our approach is then designed according to the require- ments placed by this unrestricted proof- reading protocol. In particular, the ability of the system to suggest new translations coherent with the set of potentially disjoint translation segments validated by the user. We evaluate our approach in a user- simulated setting where reference transla- tions are considered the output desired by a human expert. Results show important reductions in the number of edits in com- parison to decoupled post-editing and con- ventional prefix-based interactive transla- tion prediction. Additionally, we provide evidence that it can also reduce the cogni- tive overload reported for interactive trans- lation systems in previous user studies. 1 Introduction Research in the field of machine translation (MT) aims at developing computer systems that re- duce the effort required to generate translations, whether by assisting human translators or by di- rectly replacing them. However, most research in MT has focused on the development of fully auto- matic MT approaches. Despite that, except for a handful of very constrained domains, current au- tomatic MT technology still only achieves results that are not satisfactory in practice; automatic MT still require heavy human proof-reading to pro- duce human-quality translations. We present a new computer-assisted translation approach that integrates human translators and au- tomatic MT into a tight feedback loop. In our ap- proach, the user1 and the MT system collaborate to generate translations through a series of inter- actions. At each interaction, the system proposes its best translation for the given input sentence. If the user finds it correct, then it is accepted and the process goes on with the next input sentence. Oth- erwise, the user makes some corrections that the system takes into account to improve the proposed translation. The rationale behind this interactive translation prediction (ITP) approach is to com- bine the accuracy provided by the human expert with the efficiency of the MT system in contrast to decoupled post-editing (PE). Previous works, e.g. (Barrachina et al., 2009), have explored this paradigm; however their practical implementation limits this general proof-reading approach to a prefix-based interaction where the user is forced to correct the errors in the sentence strictly according to the reading order. Our main contribution, described in Section 3, is a new proof-reading protocol focused on pro- viding a more natural interaction between the user and the system. Specifically, …

Content truncated to 500 words