GLUE
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
GLUE Various (val)
90.8MNLI
13
GLUE 1k samples (dev)
76.27Macro Avg Score
13
GLUE 100 samples (dev)
64.24Macro Avg Score
13
GLUE
84.5MRPC Accuracy
12
GLUE (val)
85.15Average Score
12
GLUE
66.62CoLA Score
12
GLUE
51.6MNLI Score
12
GLUE
60.2CoLA Score
12
GLUE MRPC, RTE, CoLA (test, val)
88.95MRPC Accuracy
12
GLUE
97.99SST-2 Accuracy
12
GLUE
87.73CoLA Accuracy
12
GLUE
89.45MRPC Score
12
GLUE
84.3GLUE Macro Score
12
GLUE (test)
92.4QNLI Accuracy
12
GLUE adversarial (test)
72RTE Accuracy
12
GLUE (val)
83.6Aggregate Score
12
GLUE
86.33MNLI Accuracy
11
GLUE FlexiBERT search space (500 models) (aggregate)
0.711Spearman Correlation
11
GLUE (test val)
319.22MRPC Score
11
GLUE
86.86CoLA Accuracy
11
GLUE base (test dev)
71.73CoLA MCC
11
GLUE
80.5GLUE Average Score
11
GLUE
95.06SST-2 Accuracy
10
GLUE
91RTE Accuracy
10
GLUE
88.69RTE Accuracy
10