List your company

Mathematical Reasoning on GSM8K (Quality, Quality Improvement)

79.8AccuracyPLoRA
59.62464.86270.175.338Aug 4, 2025

Evaluation results

MethodLinks
PLoRABase Model=QWen-2.5-7B, Configuration=Best LoRA2025.0879.87.7
PLoRABase Model=LLAMA-3.1-8B, Configuration=Best LoRA2025.08787.5
PLoRABase Model=QWen-2.5-3B, Configuration=Best LoRA2025.0874.69.8
Default LoRABase Model=QWen-2.5-7B2025.0872.1—
PLoRABase Model=LLAMA-3.2-3B, Configuration=Best LoRA2025.0871.38
Base ModelBase Model=QWen-2.5-7B2025.0870.8—
Default LoRABase Model=LLAMA-3.1-8B2025.0870.5—
Base ModelBase Model=LLAMA-3.1-8B2025.0869.6—
Default LoRABase Model=QWen-2.5-3B2025.0864.8—
Default LoRABase Model=LLAMA-3.2-3B2025.0863.3—
Base ModelBase Model=QWen-2.5-3B2025.0861.2—
Base ModelBase Model=LLAMA-3.2-3B2025.0860.4—