Process to finetune on downstream stance detection method as discussed in paper

[codepujan]

Thanks for the great work! I was wondering what is the proper way to use it for downstream task of stance detection. Can I just load the model from Huggingface and fine tune the model on sentence pair classification task ( akin to one of MNLI style datasets)? Or is there a particular way of doing this.
Any help would be appreciated.
Thanks

[Stromlight]

Hey let's talk we can short out the thing.