Tag
This paper introduces TatBLiMP, the first linguistic minimal pairs benchmark for the Tatar language, evaluating 16 morphosyntactic phenomena across models from from-scratch Tatar models to frontier multilingual LLMs.
Presents Tatoxa, a state-of-the-art system for text detoxification in the Tatar language, outperforming existing LLMs. Introduces a new dataset and shows that cross-lingual transfer performs worse than native data.