Mathematical Reasoning on MATH (test) (Score)
42.68MATH ScorePAD
Evaluation Results
| Method | Links | |
|---|---|---|
| PADsource model=qwen3-32b, train dataset=gsm8k2026.02 | 42.68 | |
| PADsource model=qwen3-8b, train dataset=gsm8k2026.02 | 42.53 | |
| PADsource model=gemma3-27b-it, train dataset=gsm8k2026.02 | 42.17 | |
| DAsource model=qwen3-32b, train dataset=gsm8k2026.02 | 41.55 | |
| RSFTsource model=qwen3-8b, train dataset=gsm8k2026.02 | 41.39 | |
| DAsource model=qwen3-8b, train dataset=gsm8k2026.02 | 41.17 | |
| DAsource model=gemma3-27b-it, train dataset=gsm8k2026.02 | 40.94 | |
| RSFTsource model=gemma3-27b-it, train dataset=gsm8k2026.02 | 40.79 | |
| RSFTsource model=qwen3-32b, train dataset=gsm8k2026.02 | 38.74 | |
| DAsource model=qwen3-32b, train dataset=medmcqa2026.02 | 34.54 | |
| DAsource model=gemma3-27b-it, train dataset=medmcqa2026.02 | 33.23 | |
| PADsource model=gemma3-27b-it, train dataset=medmcqa2026.02 | 32.92 | |
| RSFTsource model=qwen3-32b, train dataset=medmcqa2026.02 | 32.66 | |
| PADsource model=qwen3-32b, train dataset=medmcqa2026.02 | 32.58 | |
| PADsource model=qwen3-8b, train dataset=medmcqa2026.02 | 32.27 | |
| RSFTsource model=gemma3-27b-it, train dataset=medmcqa2026.02 | 32.01 | |
| RSFTsource model=qwen3-8b, train dataset=medmcqa2026.02 | 31.7 | |
| DAsource model=qwen3-8b, train dataset=medmcqa2026.02 | 30.72 |