As Little as Possible, as Much as Necessary: Detecting Over- and Undertranslations with Contrastive Conditioning
This addresses translation quality issues for users and developers, but is incremental as it builds on existing models without introducing a new paradigm.
The paper tackles the problem of detecting omissions and additions in neural machine translation by using contrastive conditioning with off-the-shelf models, achieving accuracy comparable to supervised quality estimation methods.
Omission and addition of content is a typical issue in neural machine translation. We propose a method for detecting such phenomena with off-the-shelf translation models. Using contrastive conditioning, we compare the likelihood of a full sequence under a translation model to the likelihood of its parts, given the corresponding source or target sequence. This allows to pinpoint superfluous words in the translation and untranslated words in the source even in the absence of a reference translation. The accuracy of our method is comparable to a supervised method that requires a custom quality estimation model.