Are Larger Pretrained Language Models Uniformly Better? Comparing Performance at the Instance Level.

Ruiqi Zhong, Dhruba Ghosh, Dan Klein, Jacob Steinhardt

Browse the full ACL paper archive.