CODET: A Benchmark for Contrastive Dialectal Evaluation of Machine Translation
This work addresses the issue of dialectal robustness in machine translation for the NLP community, providing a new benchmark for evaluation.
The authors tackled the problem of neural machine translation systems' limited robustness to dialectal variations by creating CODET, a benchmark with 891 dialectal variants across twelve languages, and quantitatively demonstrated the challenges large MT models face in translating these variants.
Neural machine translation (NMT) systems exhibit limited robustness in handling source-side linguistic variations. Their performance tends to degrade when faced with even slight deviations in language usage, such as different domains or variations introduced by second-language speakers. It is intuitive to extend this observation to encompass dialectal variations as well, but the work allowing the community to evaluate MT systems on this dimension is limited. To alleviate this issue, we compile and release CODET, a contrastive dialectal benchmark encompassing 891 different variations from twelve different languages. We also quantitatively demonstrate the challenges large MT models face in effectively translating dialectal variants. All the data and code have been released.