RoBERTa-large 354M

Reported on 1 benchmark across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing1 result

Coreference ResolutiononWinograd Schema Challenge
Accuracy· 2021-04-16
73.9
best: 100 (PaLM 540B (fine-tuned))
Back to Square One: Artifact Detection, Training and Commonsense Disentanglement in the Winograd Schema arXiv:2104.08161