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

Grade School Math Word Problems on GSM8k (512 tokens)

82.6Pass@1

d-TreeRPO

70.53673.66876.879.932Dec 10, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
82.6
2025.12
82.4
2025.12
82.3
2025.12
82.1
2025.12
82
2025.12
81.9
2025.12
81.5
2025.12
78.2
2025.12
77.4
2025.12
76.8
2025.12
76.4
2025.12
76.1
2025.12
75.2
2025.12
73.7
2025.12
71