Mathematical Reasoning on Math500 (P@1, P@4)
82.25Pass@1ICT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ICTModel Size=Qwen2.5-7B2026.06 | 82.25 | 95.2 | |
| STAPOModel Size=Qwen2.5-7B2026.06 | 78.05 | 93.6 | |
| 20-EntropyModel Size=Qwen2.5-7B2026.06 | 73.3 | 93.2 | |
| GRPOModel Size=Qwen2.5-7B2026.06 | 73.25 | 93.2 | |
| BaseModel Size=Qwen2.5-7B2026.06 | 72.55 | 92.4 | |
| ICTModel Size=Qwen2.5-1.5B2026.06 | 23.55 | 37.6 | |
| STAPOModel Size=Qwen2.5-1.5B2026.06 | 18.05 | 32.8 | |
| 20-EntropyModel Size=Qwen2.5-1.5B2026.06 | 17.95 | 31.6 | |
| GRPOModel Size=Qwen2.5-1.5B2026.06 | 17.9 | 33.8 | |
| BaseModel Size=Qwen2.5-1.5B2026.06 | 17.25 | 31.4 | |
| ICTModel Size=Qwen2.5-0.5B2026.06 | 12.15 | 24.8 | |
| GRPOModel Size=Qwen2.5-0.5B2026.06 | 11.7 | 23.2 | |
| 20-EntropyModel Size=Qwen2.5-0.5B2026.06 | 11.35 | 22.4 | |
| BaseModel Size=Qwen2.5-0.5B2026.06 | 11.05 | 21.2 | |
| STAPOModel Size=Qwen2.5-0.5B2026.06 | 10.23 | 21.5 |