Does this library support the evaluation of generated content in Chinese?

I am using a self-created Chinese dataset to fine-tune a large model and would like to use these automatic evaluation criteria to validate the effectiveness of the fine-tuning. If current support for Chinese is not available, may I inquire if there are any plans to implement this in future versions? Thank you very much.