GSM8K
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
GSM8K (test)
92Accuracy
43
GSM8K (test)
89.6EM Accuracy
43
GSM8K
94.53Accuracy
42
GSM8k
73.84Accuracy (maj1@1)
42
GSM8K (test)
79.08EM Accuracy
41
GSM8K
0.951AUROC
41
GSM8k
55.95Attack Accuracy
40
GSM8K
98Original Accuracy
40
GSM8K (test)
72.5Finetune Accuracy
40
GSM8K (test)
94.39Accuracy
39
GSM8K (test)
78.92Accuracy
38
GSM8K Platinum
90.57Accuracy
37
GSM8K
3.12Avg Acceptance Length (τ)
36
GSM8K
0Total Inference Time (mm:ss)
36
GSM8K
79.1Accuracy
36
GSM8k
49.96Flexible Extract Score
35
gsm8k (test)
84.3Accuracy
35
GSM8K (test)
78.3Pass@1 Accuracy
34
GSM8K
456.9Tokens Used
33
GSM8K
7.46Spd
33
GSM8K
80.7AUROC
33
GSM8K Var
35.44Accuracy
33
GSM8K standard (test)
94.16Accuracy
33
GSM8K (test)
96.5Accuracy
33
GSM8K
3.8Speedup
32