PreQuEL: Quality Estimation of Machine Translation Outputs in Advance

TL;DR

PreQuEL introduces a novel approach to predict translation quality based solely on source text properties, reducing resource use and enhancing quality estimation by leveraging data augmentation and linguistic features.

Contribution

The paper presents the first PreQuEL task, a baseline model, and a data augmentation method that significantly improves translation quality prediction without requiring actual translations.

Findings

Data augmentation enhances prediction accuracy.

Model captures syntactic and semantic features.

Augmentation benefits general quality estimation.

Abstract

We present the task of PreQuEL, Pre-(Quality-Estimation) Learning. A PreQuEL system predicts how well a given sentence will be translated, without recourse to the actual translation, thus eschewing unnecessary resource allocation when translation quality is bound to be low. PreQuEL can be defined relative to a given MT system (e.g., some industry service) or generally relative to the state-of-the-art. From a theoretical perspective, PreQuEL places the focus on the source text, tracing properties, possibly linguistic features, that make a sentence harder to machine translate. We develop a baseline model for the task and analyze its performance. We also develop a data augmentation method (from parallel corpora), that improves results substantially. We show that this augmentation method can improve the performance of the Quality-Estimation task as well. We investigate the properties of the input text that our model is sensitive to, by testing it on challenge sets and different languages. We conclude that it is aware of syntactic and semantic distinctions, and correlates and even over-emphasizes the importance of standard NLP features.