GSM8K
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
GSM8K
92.17Accuracy
32
GSM8K
99.2ASR
32
GSM8k
86Pass@1 Accuracy
32
GSM8K
80.61AUROC
32
GSM8K OOD (test)
94.98Accuracy
32
GSM8K
97.1Avg@5 Success Rate
32
GSM8K zero-shot (test)
80.67Flexible EM
32
GSM8K 0 (test)
83.7Accuracy
32
GSM8K
82.56Accuracy
32
GSM8K
97.17Accuracy
31
GSM8K @MetaMath (test)
72.3Accuracy
31
GSM8K
87Accuracy
30
GSM8K (test)
81.5Accuracy
30
GSM8K multi-turn
89.8Average Accuracy
30
GSM8K
94.4Accuracy
30
GSM8K
97.7Ac (alpha=0.10)
30
GSM8k Right Dense
64Removal Rate
30
GSM8k Mid Dense
75Removal Rate
30
GSM8k Left Dense
82Removal Rate
30
GSM8K
72.7GSM8K Accuracy
30
GSM8K
66.6GSM8K Accuracy
30
GSM8K
92.87Pass@1
30
GSM8K
1.67Average Inference Time (s)
30
GSM8K
100True Detection Rate (TD)
30
GSM8K (test)
81.3Accuracy
30