license: apache-2.0 language: - zh
These 14k samples were randomly extracted from a large corpus of Chinese texts, containing both the original text and corresponding labels. They can be used to evaluate the quality of Chinese corpora.