PAWS-X Paraphrase Recognition cross-lingual Adversarial Dataset

The dataset contains 23,659 human-translated PAWS evaluation pairs and 296,406 machine-translated training pairs in six different languages: French, Spanish, German, Chinese, Japanese, and Korean. All translation pairs are from PAWS-Wiki Examples in .

For more details, see the accompanying paper: PAWS-X: A cross-lingual adversarial dataset for paraphrase identification