Model Evaluation for Generative AI Question 6
When using evaluation methods like BLEU or ROUGE for LLM assessment, which of the following challenges is most likely to arise if the reference dataset is limited or biased?
本练习是课程的一部分
When using evaluation methods like BLEU or ROUGE for LLM assessment, which of the following challenges is most likely to arise if the reference dataset is limited or biased?
本练习是课程的一部分