Natural Language Understanding
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
156SST-2
551
May 13, 2026
97.36SST-2 (Acc)
529
Jun 19, 2026
97.9SST-2 Accuracy
416
Feb 26, 2026
97.4SST-2
201
May 12, 2026
93.2Average Score
91
Feb 27, 2026
93.45MRPC Accuracy
90
May 29, 2026
91.3SGLUE Score
84
Feb 26, 2026
89.5Average Score (GLUE)
76
Jun 29, 2026
7,564.6QNLI
75
Jun 8, 2026
92.4BoolQ Accuracy
74
May 21, 2026
97.3SST-2 Accuracy
62
Jul 7, 2026
94MRPC Accuracy
59
Feb 26, 2026
95.18SST-2
55
Mar 26, 2026
82.4CoLA Score
54
Jul 3, 2026
94.9QNLI
47
Jun 2, 2026
71.6Accuracy
46
Jul 3, 2026
72.5Average Accuracy
41
Mar 26, 2026
69.3COLA Score
41
Feb 26, 2026
95.7SST-2
37
Feb 26, 2026
97.8SST-2 (Acc)
37
Jun 12, 2026
78.3Accuracy
36
May 21, 2026
85.6Accuracy
35
May 21, 2026
65.8Accuracy
34
May 12, 2026
95.64SST-2 Accuracy
33
Feb 26, 2026
94.5CB Accuracy
32
Mar 12, 2026