The CMNLI dataset is part of the Chinese Language Understanding Evaluation (CLUE) benchmark. It consists of two parts: XNLI and MNLI. The data comes from various sources such as fiction, telephone, travel, government, slate, etc. The original MNLI data and XNLI data were translated into Chinese and English. The original training set was retained, and the dev and test sets were created by merging and shuffling the dev set from XNLI and the matched set from MNLI, and the test set from XNLI and the mismatched set from MNLI, respectively.
Paper | Code | Results | Date | Stars |
---|