GSM8K
Benchmarks
TaskDatasetSOTA resultTrendResults
GSM8K v1 (test)
95.1Accuracy
118
GSM8K
96.8Accuracy
117
GSM8K (val)
90.8Accuracy
115
GSM8K
1Accuracy
111
gsm8k
9.43Average Generation Length (τ)
109
GSM8K
93Accuracy
105
GSM8K
96.7pass@1
102
GSM8K
96Accuracy
98
GSM8K
6.92Tau ($ au$)
97
GSM8K 8-shot
95.1Accuracy
89
GSM8k Aug
81.2Accuracy
88
GSM8K
94.62GSM8K Score
87
GSM8K Hard
61.8Accuracy
82
GSM8K 5-shot
82.2Accuracy
82
GSM8k
95.22Accuracy
80
GSM8K (test)
89.2Accuracy
79
GSM8K (test)
95.24Accuracy (ACC)
76
GSM8K
0.971Accuracy
76
GSM8k
0.738PRR
66
GSM8k
88.5ROC-AUC
66
GSM8K 200 PROBLEMS
88.6Pass@16
65
GSM8K
86.4Accuracy
63
GSM8K (test)
59.6HS
62
GSM8K
6.6Average Length
61
GSM8K
96.8Pass@1 Accuracy
61