Tag
This paper presents a benchmark for Arabic-Russian scientific translation, including a hybrid parallel corpus of 27,000 sentence pairs and fine-tuned multilingual models (mT5, NLLB, Qwen) using LoRA. The best model achieves BLEU 23.15, and the work aims to lower language barriers for scientific knowledge exchange between Arabic and Russian researchers.
Malicious packages in the Arch Linux AUR have been found inserting Russian spam into users' shell configuration files, prompting a cleanup effort by maintainers.
This paper introduces a multilingual dataset of over 100,000 movie reviews from Kazakhstan, containing Russian, Kazakh, and code-switched texts. It benchmarks classical and transformer-based models on sentiment polarity and score classification tasks.