Rule-based Morphological Inflection Improves Neural Terminology Translation

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 5902–5914 November 7–11, 2021. c⃝2021 Association for Computational Linguistics 5902 Rule-based Morphological Inflection Improves Neural Terminology Translation Weijia Xu University of Maryland weijia@cs.umd.edu Marine Carpuat University of Maryland marine@cs.umd.edu Abstract Current approaches to incorporating terminol- ogy constraints in machine translation (MT) typically assume that the constraint terms are provided in their correct morphological forms. This limits their application to real-world sce- narios where constraint terms are provided as lemmas. In this paper, we introduce a mod- ular framework for incorporating lemma con- straints in neural MT (NMT) in which linguis- tic knowledge and diverse types of NMT mod- els can be flexibly applied. It is based on a novel cross-lingual inflection module that in- flects the target lemma constraints based on the source context. We explore linguistically motivated rule-based and data-driven neural- based inflection modules and design English- German health and English-Lithuanian news test suites to evaluate them in domain adap- tation and low-resource MT settings. Results show that our rule-based inflection module helps NMT models incorporate lemma con- straints more accurately than a neural module and outperforms the existing end-to-end ap- proach with lower training costs.1 1 Introduction Incorporating terminology constraints in machine translation (MT) has proven useful to adapt translation lexical choice to new domains (Hokamp and Liu, 2017) and to improve its consistency in a document (Ture et al., 2012). In neural MT (NMT), most prior work focuses on incorporating terms in the output exactly as given, using soft (Song et al., 2019; Dinu et al., 2019; Xu and Carpuat, 2021) or hard constraints (Hokamp and Liu, 2017; Post and Vilar, 2018). These approaches are problematic when translating into morphologically rich languages where terminology should be adequately inflected in the output, while it is 1Code and test suites are released at https://github. com/Izecson/terminology-translation more natural and flexible to provide constraints as lemmas as in a dictionary. To the best of our knowledge, only one pa- per has directly addressed this problem for neural MT: (Bergmanis and Pinnis, 2021) design an NMT model trained to copy-and-inflect the terminology constraints using target lemma annotations (TLA) — TLA are synthetic training samples where the source sentence is tagged with automatically gen- erated lemma constraints. While this approach im- proves translation quality, the end-to-end training set-up prevents fast adaptation to lemmas and in- flected forms that are rare or unseen at training time. Its impact is also limited to a specific neural architecture, and it is unclear whether its benefits port to more generic sequence-to-sequence models. In this paper, we introduce a modular framework for inflecting terminology constraints in NMT. It relies on a cross-lingual inflection module that predicts the inflected form of each lemma constraint based on the source context only. The inflected lemmas can then be incorporated into NMT using any of the aforementioned constrained NMT techniques. Compared with TLA, this framework is more flexible, as it can be applied to diverse types of NMT architectures …

Content truncated to 500 words