Reasoning on MATH (Pass@1)
79Pass@1GPT-4o
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-4oModel=GPT-4o2026.05 | 79 | |
| FABSVer-7BBackbone=Qwen2.5-7B, Training=FABSVer2026.05 | 77.4 | |
| FABSVer-3BBackbone=Qwen2.5-3B, Training=FABSVer2026.05 | 67.5 | |
| FABSVer-1.5BBackbone=Qwen2.5-1.5B, Training=FABSVer2026.05 | 56.8 |