Right now the model is being finetuned on top of the pretraining explained in the t5 paper. We should try it out without pretraining.
Right now the model is being finetuned on top of the pretraining explained in the t5 paper. We should try it out without pretraining.