Right. Liu et al provide evidence against the contrast pairs being crucial (with “unmatched” meaning they just sample independently from the positive and negative contrast pair distributions):
And even the unmatched condition would still indicate better sample efficiency than prompting or finetuning:
Right. Liu et al provide evidence against the contrast pairs being crucial (with “unmatched” meaning they just sample independently from the positive and negative contrast pair distributions):
And even the unmatched condition would still indicate better sample efficiency than prompting or finetuning: