liyan06 / minicheck Goto Github PK

Hi!

How did you generate the claims in the benchmark? Did you (1) directly prompt the model to generate a claim, (2) first generate model responses and then decompose them into claims, and (3) use other ways?

Thanks!

I tried to follow the approach of MiniCheck-FT5 in the paper a few days ago, but unfortunately, it did not achieve the same good effect🤣. I also understand that this is because my finetune code was not written very well. It is not a simple text classification task. If it's convenient for you, could you please send me the finetune code of the MiniCheck-FT5.