Natural Language Understanding On Glue

Average

Results

Performance results of various models on this benchmark

Comparison Table

Model Name	Average
smart-robust-and-efficient-fine-tuning-for	89.9
bert-pre-training-of-deep-bidirectional	82.1