Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Reasoning Correctness Evaluation on GSM8K

0Average False Positive Rate

Master-RM 7B

-3.50420.14843.867.452Jul 11, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.07
0-0
2025.07
0-0
2025.07
0-0
2025.07
4.7-15
2025.07
8.5-24.9
2025.07
9.4-53.4
2025.07
12.8-25.4
2025.07
17.6-37.3
2025.07
21.5-53.6
2025.07
79.1-88.3
2025.07
83.4-89.2
2025.07
87.6-90.9