Robust Neural Machine Translation: Modeling Orthographic and Interpunctual Variation
This addresses the need for robust machine translation in informal contexts like chat and social media, but it is incremental as it builds on existing data augmentation techniques.
The paper tackled the problem of neural machine translation systems failing on non-standard orthography and punctuation by proposing a generative noise model to create adversarial examples for data augmentation. The result showed that systems trained with these examples performed nearly as well on noisy data as on clean data, with a 2-3 BLEU point improvement over baselines and 50% consistency gains in translation edit rate.
Neural machine translation systems typically are trained on curated corpora and break when faced with non-standard orthography or punctuation. Resilience to spelling mistakes and typos, however, is crucial as machine translation systems are used to translate texts of informal origins, such as chat conversations, social media posts and web pages. We propose a simple generative noise model to generate adversarial examples of ten different types. We use these to augment machine translation systems' training data and show that, when tested on noisy data, systems trained using adversarial examples perform almost as well as when translating clean data, while baseline systems' performance drops by 2-3 BLEU points. To measure the robustness and noise invariance of machine translation systems' outputs, we use the average translation edit rate between the translation of the original sentence and its noised variants. Using this measure, we show that systems trained on adversarial examples on average yield 50% consistency improvements when compared to baselines trained on clean data.