Generative Data Augmentation for Commonsense Reasoning
Open paper
Benchmark
Model
Rank
Results
coreference-resolution-on-winograd-schema-challenge
RoBERTa-large + G-DAug-Inf
#19
Accuracy: 80
common-sense-reasoning-on-winogrande
G-DAUG-Combo + RoBERTa-Large
#71
Accuracy: 71.4
Rank counts only results with a code link.