MWS AI

How to Translate Mixed Russian and Kazakh Using Synthetic Data

Researchers from MWS AI and several universities have developed a method for generating synthetic datasets to translate mixed Russian-Kazakh code-switched language. Their approach, based on SimAlign alignment and the NLLB-3.3B model, outperformed commercial systems in human evaluation, although automatic metrics show the opposite.

Meta

How to Translate Mixed Russian-Kazakh and Not Go Crazy

Researchers from MWS AI and universities proposed a method for generating a synthetic dataset for machine translation of mixed Russian-Kazakh text (code-switching). Their model, trained on synthetic data, outperformed commercial systems in human evaluation, though it lagged behind them in automatic metrics.