Mathematical Reasoning on GSM8K original and paraphrased (train)
33.4S_refQwen-Math
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Qwen-MathData Variant=Original, Metric Type=ACC(%)2026.05 | 33.4 | 45.8 | — | — | — | 1 | |
| Qwen-MathData Variant=Paraphrased, Metric Type=ACC(%)2026.05 | 33.4 | 42 | — | — | — | 0.996 | |
| Qwen-MathData Variant=Original, Metric Type=Con(%)2026.05 | 33.2 | — | 45.8 | — | — | 1 | |
| Qwen-MathData Variant=Paraphrased, Metric Type=Con(%)2026.05 | 33.2 | — | 41.8 | — | — | 0.997 | |
| DeepSeek-MathData Variant=Original, Metric Type=ACC(%)2026.05 | 22.2 | 29.8 | — | — | — | 0.989 | |
| DeepSeek-MathData Variant=Paraphrased, Metric Type=ACC(%)2026.05 | 22.2 | 27.6 | — | — | — | 0.951 | |
| DeepSeek-MathData Variant=Original, Metric Type=Con(%)2026.05 | 21.6 | — | 30 | — | — | 0.997 | |
| DeepSeek-MathData Variant=Paraphrased, Metric Type=Con(%)2026.05 | 21.6 | — | 26.6 | — | — | 0.92 | |
| Qwen-MathData Variant=Original, Metric Type=P_first2026.05 | 0.488 | — | — | 0.532 | — | 0.962 | |
| Qwen-MathData Variant=Paraphrased, Metric Type=P_first2026.05 | 0.488 | — | — | 0.532 | — | 0.94 | |
| Qwen-MathData Variant=Original, Metric Type=P_all2026.05 | 0.412 | — | — | — | 0.511 | 0.998 | |
| Qwen-MathData Variant=Paraphrased, Metric Type=P_all2026.05 | 0.412 | — | — | — | 0.498 | 0.998 | |
| DeepSeek-MathData Variant=Original, Metric Type=P_first2026.05 | 0.38 | — | — | 0.463 | — | 0.998 | |
| DeepSeek-MathData Variant=Paraphrased, Metric Type=P_first2026.05 | 0.38 | — | — | 0.459 | — | 0.998 | |
| DeepSeek-MathData Variant=Original, Metric Type=P_all2026.05 | 0.285 | — | — | — | 0.395 | 0.998 | |
| DeepSeek-MathData Variant=Paraphrased, Metric Type=P_all2026.05 | 0.285 | — | — | — | 0.381 | 0.998 |