List your company

Mathematical Reasoning on GSM8K (FRS, pass@1 accuracy)

98.9FRSDS-R1-7B
38.89254.47170.0585.629Apr 13, 2026

Evaluation results

MethodLinks
DS-R1-7BFiltering rate=10%, Decoding strategy=greedy2026.0498.9—
DS-R1-1.5BFiltering rate=10%, Decoding strategy=greedy2026.0495.1—
Phi-4Filtering rate=10%, Decoding strategy=greedy2026.0491.5—
Qwen3-4BFiltering rate=10%, Decoding strategy=greedy2026.0490.9—
Qwen2.5-MathFiltering rate=10%, Decoding strategy=greedy2026.0490.8—
Phi-4-Reas.Filtering rate=10%, Decoding strategy=greedy2026.0484.5—
LLaMA-3.1-8BFiltering rate=10%, Decoding strategy=greedy2026.0479.9—
Qwen2.5-7BFiltering rate=10%, Decoding strategy=greedy2026.0465.6—
Gemma-7BFiltering rate=10%, Decoding strategy=greedy2026.0441.2—