10,708 Hits in 7.8 sec

Online adaptation to post-edits for phrase-based statistical machine translation

Nicola Bertoldi, Patrick Simianer, Mauro Cettolo, Katharina Wäschle, Marcello Federico, Stefan Riezler
2014 Machine Translation  
We present an efficient online learning framework for adapting all modules of a phrase-based statistical machine translation system to post-edited translations.  ...  We use a constrained search technique to extract new phrase-translations from post-edits without the need of re-alignments, and to extract phrase pair features for discriminative training without the need  ...  Acknowledgments FBK researchers were supported by the MateCat project, funded by the EC under FP7; researchers at Heidelberg University by DFG grant "Cross-language Learning-to-Rank for Patent Retrieval  ... 
doi:10.1007/s10590-014-9159-7 fatcat:y6dkepdocrc75bvk7rzhhzsvwq

Learning from Post-Editing: Online Model Adaptation for Statistical Machine Translation

Michael Denkowski, Chris Dyer, Alon Lavie
2014 Proceedings of the 14th Conference of the European Chapter of the Association for Computational Linguistics  
We propose and evaluate three computationally efficient online methods for updating statistical MT systems in a scenario where post-edited MT output is constantly being returned to the system: (1) adding  ...  Using machine translation output as a starting point for human translation has become an increasingly common application of MT.  ...  Conclusion Casting machine translation for post-editing as an online learning task, we have presented three methods for incremental model adaptation: adding data to the indexed bitext from which grammars  ... 
doi:10.3115/v1/e14-1042 dblp:conf/eacl/DenkowskiDL14 fatcat:fdyizdro7ffwhkx6vrl7biqm2i

Translating without in-domain corpus: Machine translation post-editing with online learning techniques

Antonio L. Lagarda, Daniel Ortiz-Martínez, Vicent Alabau, Francisco Casacuberta
2015 Computer Speech and Language  
In order to solve this problem, we propose the application of automatic post-editing in an online learning framework.  ...  This automatic post-editing module learns to make its corrections from user feedback in real time by means of online learning techniques.  ...  Framework In this section we describe the theoretical foundations of our proposal, where we apply online learning techniques to an automatic post-editing task based on statistical machine translation.  ... 
doi:10.1016/j.csl.2014.10.004 fatcat:sanhtrtosfeifcyht7s4mnmi6q

Cache-based Online Adaptation for Machine Translation Enhanced Computer Assisted Translation

Nicola Bertoldi, Mauro Cettolo, Marcello Federico
2013 Machine Translation Summit  
One of them, addressed in this work, is how to dynamically adapt phrase-based statistical MT from user post-editing.  ...  By casting the problem in the online machine learning paradigm, we propose a cache-based adaptation technique method that dynamically stores target n-gram and phrase-pair features used by the translator  ...  LM Adaptation Similarly to the local translation model, a local language model is built to reward the n-grams found in post-edited translation.  ... 
dblp:conf/mtsummit/BertoldiCF13 fatcat:ksoa5j4yfnezxf6rbabsco63t4

Online Learning Approaches in Computer Assisted Translation

Prashant Mathur, Mauro Cettolo, Marcello Federico
2013 Conference on Machine Translation  
We present a novel online learning approach for statistical machine translation tailored to the computer assisted translation scenario.  ...  Our results show that our online adaptation technique outperforms the static phrase based statistical machine translation system by 6 BLEU points absolute, and a standard incremental adaptation approach  ...  Recent research has led to the integration of CAT tools with statistical machine translation (SMT) engines.  ... 
dblp:conf/wmt/MathurCF13 fatcat:wak4updsmjgmlfo2vw4qdnr22u

A User-Study on Online Adaptation of Neural Machine Translation to Human Post-Edits [article]

Sariya Karimova, Patrick Simianer, Stefan Riezler (Heidelberg University)
2018 arXiv   pre-print
We present the first user study on online adaptation of NMT to user post-edits in the domain of patent translation.  ...  Our experimental results show a significant reduction of human post-editing effort due to online adaptation in NMT according to several evaluation metrics, including hTER, hBLEU, and KSMR.  ...  User-studies on the benefits of machine learning for adaptation of translation systems to human post-edits are rare, and to the best of our knowledge, such studies have so far been restricted to phrase-based  ... 
arXiv:1712.04853v3 fatcat:igmcvmmvond5hdh5wezizw4r4m

Real Time Adaptive Machine Translation for Post-Editing with cdec and TransCenter

Michael Denkowski, Alon Lavie, Isabel Lacruz, Chris Dyer
2014 Proceedings of the EACL 2014 Workshop on Humans and Computer-assisted Translation  
This paper describes cdec Realtime, a framework for building adaptive MT systems that learn from post-editor feedback, and TransCenter, a web-based translation interface that connects users to Realtime  ...  Using machine translation output as a starting point for human translation has recently gained traction in the translation community.  ...  The MateCat 8 project (Cattelan, 2013) specifically aims to integrate machine translation (including online model adaptation and translation quality estimation) into a web-based CAT tool.  ... 
doi:10.3115/v1/w14-0311 dblp:conf/eacl/DenkowskiLLD14 fatcat:ngofbymmajazxa6asd7c73v6ly

Instance Selection for Online Automatic Post-Editing in a multi-domain scenario

Rajen Chatterjee, Mihael Arcan, Matteo Negri, Marco Turchi
2016 Conference of the Association for Machine Translation in the Americas  
In recent years, several end-to-end online translation systems have been proposed to successfully incorporate human post-editing feedback in the translation workflow.  ...  The performance of these systems in a multi-domain translation environment (involving different text genres, post-editing styles, machine translation systems) within the automatic post-editing (APE) task  ...  These systems mainly follow the phrase-based machine translation approach where the MT outputs (with optionally the source sentence) are used as the source language corpus and the post-edits are used as  ... 
dblp:conf/amta/ChatterjeeANT16 fatcat:7xxywap24bgitghvqyivi47l6u

A Novel Rule Refinement Method for SMT through Simulated Post-Editing [chapter]

Sitong Yang, Heng Yu, Qun Liu
2014 Communications in Computer and Information Science  
Post-editing has been successfully applied to correct the output of MT systems to generate better translation, but as a downstream task its positive feedback to MT has not been well studied.  ...  In this paper, we present a novel rule refinement method which uses Simulated Post-Editing (SiPE) to capture the errors made by the MT systems and generates refined translation rules.  ...  Acknowledgement We thank the three anonymous reviewers for helpful suggestions. The authors were supported by CAS Action Plan for the Development of Western China (No.  ... 
doi:10.1007/978-3-662-45924-9_11 fatcat:bmzzv6uvsvf7rlnau3i2lle53i

Demonstration of a Neural Machine Translation System with Online Learning for Translators

Miguel Domingo, Mercedes García-Martínez, Amando Estela Pastor, Laurent Bié, Alexander Helle, Álvaro Peris, Francisco Casacuberta, Manuel Herranz Pérez
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: System Demonstrations  
Our objective was to save post-editing effort as the machine is continuously learning from human choices and adapting the models to a specific domain or user style.  ...  We introduce a demonstration of our system, which implements online learning for neural machine translation in a production environment.  ...  Acknowledgments The research leading to these results has received funding from the Spanish Centre for Technological and Industrial Development (Centro para el Desarrollo Tecnológico Industrial) (CDTI)  ... 
doi:10.18653/v1/p19-3012 dblp:conf/acl/DomingoGEBHPCH19 fatcat:jbl2y4mdozdkpbux2pttznc5ie

Findings of the 2016 Conference on Machine Translation

Ondřej Bojar, Rajen Chatterjee, Christian Federmann, Yvette Graham, Barry Haddow, Matthias Huck, Antonio Jimeno Yepes, Philipp Koehn, Varvara Logacheva, Christof Monz, Matteo Negri, Aurelie Neveol (+9 others)
2016 Proceedings of the First Conference on Machine Translation: Volume 2, Shared Task Papers  
This year, 102 MT systems from 24 institutions (plus 36 anonymized online systems) were submitted to the 12 translation directions in the news translation task.  ...  The automatic post-editing task had a total of 6 teams, submitting 11 entries. 1 2  ...  Four online statistical machine translation systems Table 2 : Participants in the shared translation task. Not all teams participated in all language pairs.  ... 
doi:10.18653/v1/w16-2301 dblp:conf/wmt/BojarCFGHHJKLMN16 fatcat:jqwhstyem5ahrns5v4tju233ay

Online Neural Automatic Post-editing for Neural Machine Translation [chapter]

Matteo Negri, Marco Turchi, Nicola Bertoldi, Marcello Federico
2018 Proceedings of the Fifth Italian Conference on Computational Linguistics CLiC-it 2018  
Machine learning from user corrections is key to the industrial deployment of machine translation (MT).  ...  We present experimental results of APE on English-Italian MT by simulating human post-edits with human reference translations, and by applying online APE on MT outputs of increasing quality.  ...  Online learning capabilities are crucial (both for APE and MT) in computer-assisted translation scenarios where professional translators operate on suggestions provided by machines.  ... 
doi:10.4000/books.aaccademia.3534 fatcat:glvph3cuz5b6hkk3jksc6ul3xa

Preliminary Experiments on Using Users' Post-Editions to Enhance a SMT System Oracle-based Training for Phrase-based Statistical Machine Translation

Ankit Srivastava, Yanjun Ma, Andy Way
2011 European Association for Machine Translation Conferences/Workshops  
The aim is to develop an interactive and iterative translation system able to adapt and improve itself rapidly using user's feedbacks (collected as translation posteditions).  ...  The results show that, despite the difficulty to measure the system improvement with automatic metrics (given the small size of our collected post-edition corpus) a small-sized subjective analysis reveals  ...  Analysis for Adaptive Statistical Translation) organizes the collection and analysis of users' feedbacks from the online Reverso translation service in order to develop techniques to exploit them (Déchelotte  ... 
dblp:conf/eamt/SrivastavaMW11 fatcat:3cr2k3jwlnftvgoyn2jycvfnwq

Online Automatic Post-editing for MT in a Multi-Domain Translation Environment

Rajen Chatterjee, Gebremedhen Gebremelak, Matteo Negri, Marco Turchi
2017 Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers  
Automatic post-editing (APE) for machine translation (MT) aims to fix recurrent errors made by the MT decoder by learning from correction examples.  ...  Ideally, to be integrated in a real professional translation workflow (e.g. to play a role in computerassisted translation framework), APE tools should be flexible enough to cope with continuous streams  ...  Introduction Automatic post-editing (APE) systems for machine translation (MT) aim to correct the errors present in a machine-translated text before showing it to the user, thereby reducing human workload  ... 
doi:10.18653/v1/e17-1050 dblp:conf/eacl/NegriTCG17 fatcat:rr42jhtamjcpreomh553lmutvu

Findings of the 2017 Conference on Machine Translation (WMT17)

Ondřej Bojar, Rajen Chatterjee, Christian Federmann, Yvette Graham, Barry Haddow, Shujian Huang, Matthias Huck, Philipp Koehn, Qun Liu, Varvara Logacheva, Christof Monz, Matteo Negri (+4 others)
2017 Proceedings of the Second Conference on Machine Translation  
This year, 102 MT systems from 24 institutions (plus 36 anonymized online systems) were submitted to the 12 translation directions in the news translation task.  ...  , tuning, run-time estimation of MT quality), and an automatic post-editing task and bilingual document alignment task.  ...  The APE task organizers would also like to thank Jan Niehues for training the KIT system used to produce the MT output, Text&Form for producing the manual post-edits, and the annotators involved in the  ... 
doi:10.18653/v1/w17-4717 dblp:conf/wmt/BojarCFGHHHKLLM17 fatcat:yawnu6qao5ehpnfqlzdgmxygtm
« Previous Showing results 1 — 15 out of 10,708 results