Armenian Paraphrase Detection Corpus Dataset

This dataset contains 2,360 paraphrases in Armenian that can be used for paraphrase detection. The dataset is constructed by back-translating sentences from Armenian to English twice, and manually filtering the result.

本数据集包含2,360条亚美尼亚语的释义句，可用于释义检测。该数据集通过两次将亚美尼亚语句子翻译成英语并反向翻译回亚美尼亚语的方式构建，随后进行人工筛选。