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

Advanced Mathematical Reasoning on Math500 (512 tokens)

46.3Pass@1 Accuracy

d-TreeRPO

34.3437.44540.5543.655Dec 10, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
46.3
2025.12
44.1
2025.12
43.4
2025.12
43.4
2025.12
42.2
2025.12
41.2
2025.12
40.4
2025.12
39.1
2025.12
39
2025.12
39
2025.12
38.9
2025.12
38.5
2025.12
38.4
2025.12
36.2
2025.12
34.8