GSM8K
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
GSM8K
93.83Mean Accuracy
3
GSM8K (val)
0.945AUROC
3
GSM8K
95.6Accuracy
3
GSM8K
99.9TPR
3
GSM8K
79.37Deductive Accuracy
3
GSM8K
24.8Perplexity (PPL)
3
GSM8K Insufficient
49.3Success Rate (SR)
3
GSM8K
0.955p0 Score
3
GSM8K
44.5Detection Accuracy
3
GSM8K
69.95TPS
3
GSM8k
62.55GSM8k Zero-shot Accuracy
3
GSM8K
95CSR
3
GSM8K
23.45Perplexity (PPL)
3
GSM8K (test)
0.852Best-layer ROC-AUC
3
GSM8K
95.3Strict Accuracy
3
GSM8K 2021
82.6Peak Accuracy
3
GSM8K All Non-Easy
29.7Recovery Rate
3
GSM8K Med. + Hard
27.6Recovery Rate (%)
3
GSM8K
72.4AUROC
3
GSM8k
78.09Accuracy
3
GSM8K DE 8-shot (test)
47.5Accuracy
3
GSM8K
55Accuracy (GSM8K)
3
GSM8K
94Chain Precision
3
GSM8K (test)
143.93TPS
3
GSM8K 5
79Performance Score
3