Utilize Transformers for translating Wikipedia category names
Hoang-Thang Ta, Quoc Thang La
TL;DR
This work tackles translating Wikipedia category names from English to Vietnamese to streamline category creation in a low-resource setting. It trains and compares small to mid-size Transformer-based models (BART-base, T5-small, OPUS-MT-en-vi) on a 15,000-pair English–Vietnamese Wikidata-derived dataset, finding OPUS-MT-en-vi provides the best BLEU score (0.73) with the smallest footprint. The study demonstrates that resource-efficient Transformer models can perform competitive category-name translation, while also highlighting limitations such as dataset size and lack of human baseline evaluation. The results suggest practical applicability for editors and point to avenues for scaling to larger, multilingual models and broader category coverage.
Abstract
On Wikipedia, articles are categorized to aid readers in navigating content efficiently. The manual creation of new categories can be laborious and time-intensive. To tackle this issue, we built language models to translate Wikipedia categories from English to Vietnamese with a dataset containing 15,000 English-Vietnamese category pairs. Subsequently, small to medium-scale Transformer pre-trained models with a sequence-to-sequence architecture were fine-tuned for category translation. The experiments revealed that OPUS-MT-en-vi surpassed other models, attaining the highest performance with a BLEU score of 0.73, despite its smaller model storage. We expect our paper to be an alternative solution for translation tasks with limited computer resources.