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

Mathematical Reasoning on GSM8K, MATH 500, and TheoremQA

68.14Math Score

A-CHORD

Updated 1mo ago

Evaluation Results

Method	Links
A-CHORD 2025.09		68.14
CHORD 2025.09		67.67
GRPO 2025.09		66.52
A-GRPO 2025.09		65.12
A-SFT → A-GRPO 2025.09		63.65
SFT → A-GRPO 2025.09		60.53
Qwen3-4B 2025.09		59.47
SFT → GRPO 2025.09		59.4
SFT 2025.09		56.92
A-SFT 2025.09		56.65
A-SFT → A-GRPO 2025.09		43.72
Qwen3-0.6B 2025.09		42.29
GRPO 2025.09		41.49
A-GRPO 2025.09		41.04
A-SFT 2025.09		40.14
SFT → A-GRPO 2025.09		39.95
SFT → GRPO 2025.09		39.57
SFT 2025.09		39.56
A-CHORD 2025.09		39.45
CHORD 2025.09		38.52