Mathematical Reasoning on GSM8K (Quality, Quality Improvement)
79.8AccuracyPLoRA
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PLoRABase Model=QWen-2.5-7B, Configuration=Best LoRA2025.08 | 79.8 | 7.7 | |
| PLoRABase Model=LLAMA-3.1-8B, Configuration=Best LoRA2025.08 | 78 | 7.5 | |
| PLoRABase Model=QWen-2.5-3B, Configuration=Best LoRA2025.08 | 74.6 | 9.8 | |
| Default LoRABase Model=QWen-2.5-7B2025.08 | 72.1 | — | |
| PLoRABase Model=LLAMA-3.2-3B, Configuration=Best LoRA2025.08 | 71.3 | 8 | |
| Base ModelBase Model=QWen-2.5-7B2025.08 | 70.8 | — | |
| Default LoRABase Model=LLAMA-3.1-8B2025.08 | 70.5 | — | |
| Base ModelBase Model=LLAMA-3.1-8B2025.08 | 69.6 | — | |
| Default LoRABase Model=QWen-2.5-3B2025.08 | 64.8 | — | |
| Default LoRABase Model=LLAMA-3.2-3B2025.08 | 63.3 | — | |
| Base ModelBase Model=QWen-2.5-3B2025.08 | 61.2 | — | |
| Base ModelBase Model=LLAMA-3.2-3B2025.08 | 60.4 | — |