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

Mathematical Reasoning on GSM-Infinite (8K)

22.9Accuracy

TriangleMix

2.627.88513.1518.415Jul 29, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.07
22.9
2025.07
21.6
2025.07
21.3
2025.07
21.3
2025.07
21
2025.07
20.4
2025.07
18.6
2025.07
17.7
2025.07
8.6
2025.07
7.6
2025.07
7.3
2025.07
7
2025.07
6.9
2025.07
6.5
2025.07
6.3
2025.07
6.3
2025.07
5.3
2025.07
3.8
2025.07
3.8
2025.07
3.7
2025.07
3.5
2025.07
3.4
2025.07
3.4
2025.07
3.4