Christian Hardmeier

Pronoun-Focused MT and Cross-Lingual Pronoun Prediction: Findings of the 2015 DiscoMT Shared Task on Pronoun Translation

We describe the design, the evaluation setup, and the results of the DiscoMT 2015 shared task, which included two subtasks, relevant to both the machine translation (MT) and the discourse communities: (i) pronoun-focused translation, a practical MT task, and (ii) cross-lingual pronoun prediction, a classification task that requires no specific MT expertise and is interesting as a machine learning task in its own right. We focused on the English‐French language pair, for which MT output is generally of high quality, but has visible issues with pronoun translation due to differences in the pronoun systems of the two languages. Six groups participated in the pronoun-focused translation task and eight groups in the cross-lingual pronoun prediction task.

Part-of-Speech Driven Cross-Lingual Pronoun Prediction with Feed-Forward Neural Networks

For some language pairs, pronoun translation is a discourse-driven task which requires information that lies beyond its local context. This motivates the task of predicting the correct pronoun give ...

Findings of the 2016 WMT Shared Task on Cross-lingual Pronoun Prediction

We describe the design, the evaluation setup, and the results of the 2016 WMT shared task on cross-lingual pronoun prediction. This is a classification task in which participants are asked to provi ...

Anaphora Models and Reordering for Phrase-Based SMT

We describe the Uppsala University systems for WMT14. We look at the integration of a model for translating pronominal anaphora and a syntactic dependency projection model for English‐French. Furthermore, we investigate post-ordering and tunable POS distortion models for English‐ German.

It-disambiguation and source-aware language models for cross-lingual pronoun prediction.

We present our systems for the WMT 2016 shared task on cross-lingual pronoun prediction. The main contribution is a classifier used to determine whether an instance of the ambiguous English pronoun “it” functions as an anaphoric, pleonastic or event reference pronoun. For the English-to-French task the classifier is incorporated in an extended baseline, which takes the form of a source-aware language model. An implementation of the sourceaware language model is also provided for each of the remaining language pairs.

Pronoun Prediction with Latent Anaphora Resolution

This paper describes the UU-Hardmeier submissions to the WMT 2016 shared task on cross-lingual pronoun prediction. Our model is a system combination of two different approaches, one based on a neural network with latent anaphora resolution and the other one on ann-gram model with an additional dependency on the source pronoun. The combination of the two models results in an improvement over each individual system, but it appears that the contribution of the neural network is more likely due to its context modelling capacities than to the anaphora resolution subnetwork.

A Document-Level SMT System with Integrated Pronoun Prediction

This paper describes one of Uppsala University’s submissions to the pronoun-focused machine translation (MT) shared task at DiscoMT 2015. The system is based on phrase-based statistical MT implemen ...

On Statistical Machine Translation and Translation Theory

The translation process in statistical machine translation (SMT) is shaped by technical constraints and engineering considerations. SMT explicitly models translation as search for a target-language equivalent of the input text. This perspective on translation had wide currency in mid-20th century translation studies, but has since been superseded by approaches arguing for a more complex relation between source and target text. In this paper, we show how traditional assumptions of translational equivalence are embodied in SMT through the concepts of word alignment and domain and discuss some limitations arising from the word-level/corpus-level dichotomy inherent in these concepts.

Kaibao Hu: Introducing corpus-based translation studies

Introducing Corpus-Based Translation Studies by Kaibao Hu, published by Springer in 2016, is an English translation of a 2011 book originally published in Chinese by Shanghai Jiao Tong University Press. The book introduces corpus-based translation studies as a new paradigm that parallels the shift from introspective to corpus-based methods in linguistics. In a metaphor that recurs throughout the text, corpus-based translation studies are described as a “marriage”betweendescriptive translation studies and corpus linguistics. The book is divided into nine chapters. The first two address the definition of corpus-based translation studies and the corpora and tools commonly used in this field. Six chapters outline the use of corpus-based methods for different research questions in translation studies. The last chapter contains an outlook and discusses some of the problems and limitations of the approach. The first chapter, Introduction, introduces corpus-based translation studies and contrasts it with conventional translation studies. As a foil for this descriptive, corpusbased approach, the author describes traditional methods as prescriptive, relying on intuition and anecdotal evidence and characterised by the primacy of the source text and the concept of translational equivalence. This suggests, perhaps not entirely uncontroversially, that the corpus-based method is the only method appropriate to modern translation studies after the cultural turn of the late twentieth century. However, the chapter does not discuss the development of the discipline at large and instead contains a lengthy and very dense history of corpus-based translation studies, with sections devoted to such special interests as conferences and workshops on corpus-based translation studies until about 2003 and corpus-based translation studies inChina (including a two-page list of research projects with national Chinese funding).