Abstract:Incorporating extra-textual context such as film metadata into the machine translation (MT) pipeline can enhance translation quality, as indicated by automatic evaluation in recent work. However, the positive impact of such systems in industry remains unproven. We report on an industrial case study carried out to investigate the benefit of MT in a professional scenario of translating TV subtitles with a focus on how leveraging extra-textual context impacts post-editing. We found that post-editors marked significantly fewer context-related errors when correcting the outputs of MTCue, the context-aware model, as opposed to non-contextual models. We also present the results of a survey of the employed post-editors, which highlights contextual inadequacy as a significant gap consistently observed in MT. Our findings strengthen the motivation for further work within fully contextual MT.
Abstract:Efficient utilisation of both intra- and extra-textual context remains one of the critical gaps between machine and human translation. Existing research has primarily focused on providing individual, well-defined types of context in translation, such as the surrounding text or discrete external variables like the speaker's gender. This work introduces MTCue, a novel neural machine translation (NMT) framework that interprets all context (including discrete variables) as text. MTCue learns an abstract representation of context, enabling transferability across different data settings and leveraging similar attributes in low-resource scenarios. With a focus on a dialogue domain with access to document and metadata context, we extensively evaluate MTCue in four language pairs in both translation directions. Our framework demonstrates significant improvements in translation quality over a parameter-matched non-contextual baseline, as measured by BLEU (+0.88) and Comet (+1.58). Moreover, MTCue significantly outperforms a "tagging" baseline at translating English text. Analysis reveals that the context encoder of MTCue learns a representation space that organises context based on specific attributes, such as formality, enabling effective zero-shot control. Pre-training on context embeddings also improves MTCue's few-shot performance compared to the "tagging" baseline. Finally, an ablation study conducted on model components and contextual variables further supports the robustness of MTCue for context-based NMT.
Abstract:Personalisation of language models for dialogue sensitises them to better capture the speaking patterns of people of specific characteristics, and/or in specific environments. However, rich character annotations are difficult to come by and to successfully leverage. In this work, we release and describe a novel set of manual annotations for 863 speakers from the popular Cornell Movie Dialog Corpus, including features like characteristic quotes and character descriptions, and a set of six automatically extracted metadata for over 95% of the featured films. We perform extensive experiments on two corpora and show that such annotations can be effectively used to personalise language models, reducing perplexity by up to 8.5%. Our method can be applied even zero-shot for speakers for whom no prior training data is available, by relying on combinations of characters' demographic characteristics. Since collecting such metadata is costly, we also contribute a cost-benefit analysis to highlight which annotations were most cost-effective relative to the reduction in perplexity.