Using Coreference Links to Improve Spanish-to-English Machine Translation

Abstract

In this paper, we present a proof-ofconcept of a coreference-aware decoder for document-level machine translation. We consider that better translations should have coreference links that are closer to those in the source text, and implement this criterion in two ways. First, we define a similarity measure between source and target coreference structures, by projecting the target ones onto the source ones, and then reusing existing monolingual coreference metrics. Based on this similarity measure, we re-rank the translation hypotheses of a baseline MT system for each sentence. Alternatively, to address the lack of diversity of mentions among the MT hypotheses, we focus on mention pairs and integrate their coreference scores with MT ones, resulting in post-editing decisions. Experiments with Spanish-to-English MT on the AnCora-ES corpus show that our second approach yields a substantial increase in the accuracy of pronoun translation, while BLEU scores remain constant.

Document-Level Neural Machine Translation with Hierarchical Attention Networks

“…Statistical Machine Translation (SMT) Initial studies were based on cache memories (Tiede- mann, 2010; Gong et al, 2011). However, most of the work explicitly models discourse phenomena (Sim Smith, 2017) such as lexical cohesion (Meyer and Popescu-Belis, 2012;Xiong et al, 2013;Loáiciga and Grisot, 2016;Pu et al, 2017;Mascarell, 2017), coherence (Born et al, 2017), and coreference (Rios Gonzales and Tuggener, 2017;Miculicich Werlen and Popescu-Belis, 2017a). Hardmeier et al (2013) introduced the document-level SMT paradigm.…”

Validation of an Automatic Metric for the Accuracy of Pronoun Translation (APT)

“…APT has been used for evaluating Spanishto-English pronoun translation Luong et al, 2017;Miculicich Werlen and Popescu-Belis, 2017), showing that it can be adapted to other language pairs.…”

“…APT has been used for evaluating Spanishto-English pronoun translation Luong et al, 2017;Miculicich Werlen and Popescu-Belis, 2017), showing that it can be adapted to other language pairs. While it is not likely that large shared tasks such as the WMT Metrics Task (Stanojević et al, 2015) can be designed for assessing pronoun evaluation metrics only, we believe that, in the future, the availability of larger amounts of human ratings from new shared tasks on pronoun translation will offer new opportunities to confirm the accuracy of APT and possibly to tune its parameters for an even increased correlation.…”

Word-Level Coreference Resolution

“…Coreference resolution is used in various natural language processing pipelines, such as machine translation (Ohtani et al, 2019;Miculicich Werlen and Popescu-Belis, 2017), question answering (Dhingra et al, 2018) and text simplification (Wilkens et al, 2020). As a building block of larger systems itself, it is crucial that coreference resolution models aim to not only achieve higher benchmark scores but be time and memory efficient as well.…”