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

Multi-token generation on GSM8K

68.8Accuracy

IA2 → SFT

44.46450.78257.163.418Sep 26, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.09
68.8
2025.09
68.4
2025.09
66.2
2025.09
64.5
2025.09
45.4