Bilingual Synchronization: Restoring Translational Relationships with Editing Operations
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 30 of its 67 references that are in the map. First posted on arXiv 24 Oct 2022.
Builds on
30 of 67 references.
References · 30
- 2017Attention Is All You Needcited by 60,361
- 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,548
- 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,918
- 2018A Call for Clarity in Reporting BLEU Scorescited by 1,799
- 2016Enriching Word Vectors with Subword Informationcited by 1,583
- 2019fairseq: A Fast, Extensible Toolkit for Sequence Modelingcited by 1,488
- 2019Cross-lingual Language Model Pretrainingcited by 1,090
- 2020COMET: A Neural Framework for MT Evaluationcited by 1,033
- 2016Using the Output Embedding to Improve Language Modelscited by 272
- 2017Lexically Constrained Decoding for Sequence Generation Using Grid Beam Searchcited by 237
- 2020Nearest Neighbor Machine Translationcited by 209
- 2019Levenshtein Transformercited by 199
18 more
- 2018Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinementcited by 195
- 2018Fast Lexically Constrained Decoding with Dynamic Beam Allocation for Neural Machine Translationcited by 182
- 2016Tying Word Vectors and Word Classifiers: A Loss Framework for Language Modelingcited by 92
- 2021Neural Machine Translation with Monolingual Translation Memorycited by 61
- 2020Lexically Constrained Neural Machine Translation with Levenshtein Transformercited by 45
- 2018WikiAtomicEdits: A Multilingual Corpus of Wikipedia Edits for Modeling Language and Discoursecited by 26
- 2020EDITOR: an Edit-Based Transformer with Repositioning for Neural Machine Translation with Soft Lexical Constraintscited by 24
- 2021TranSmart: A Practical Interactive Machine Translation Systemcited by 17
- 2016Log-linear Combinations of Monolingual and Bilingual Neural Machine Translation Models for Automatic Post-Editingcited by 13
- 2018MS-UEdin Submission to the WMT2018 APE Shared Task: Dual-Source Transformer for Automatic Post-Editingcited by 13
- 2021IntelliCAT: Intelligent Machine Translation Post-Editing with Quality Estimation and Translation Suggestioncited by 11
- 2022Non-Autoregressive Machine Translation: It's Not as Fast as it Seemscited by 11
- 2017Enabling Multi-Source Neural Machine Translation By Concatenating Source Sentences In Multiple Languagescited by 10
- 2019Unbabel's Submission to the WMT2019 APE Shared Task: BERT-based Encoder-Decoder for Automatic Post-Editingcited by 8
- 2021WeTS: A Benchmark for Translation Suggestioncited by 8
- 2017Ensembling Factored Neural Machine Translation Models for Automatic Post-Editing and Quality Estimationcited by 5
- 2016CUNI System for WMT16 Automatic Post-Editing and Multimodal Translation Taskscited by 4
- 2017LIG-CRIStAL System for the WMT17 Automatic Post-Editing Taskcited by 1
Built on by
Citing papers per year: 2023: 2.
2 papers.