Mathematical Reasoning on AIME 24 (P@1, P@4)
10.83P@1ICT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ICTModel Size=Qwen2.5-7B2026.06 | 10.83 | 23.33 | |
| BaseModel Size=Qwen2.5-7B2026.06 | 10 | 20 | |
| STAPOModel Size=Qwen2.5-7B2026.06 | 10 | 20 | |
| 20-EntropyModel Size=Qwen2.5-7B2026.06 | 6.67 | 13.33 | |
| GRPOModel Size=Qwen2.5-7B2026.06 | 5 | 16.67 | |
| ICTModel Size=Qwen2.5-1.5B2026.06 | 2.84 | 13.64 | |
| STAPOModel Size=Qwen2.5-1.5B2026.06 | 1.71 | 4.55 | |
| BaseModel Size=Qwen2.5-1.5B2026.06 | 1.42 | 6.82 | |
| STAPOModel Size=Qwen2.5-0.5B2026.06 | 0.57 | 2.27 | |
| ICTModel Size=Qwen2.5-0.5B2026.06 | 0.57 | 2.27 | |
| 20-EntropyModel Size=Qwen2.5-1.5B2026.06 | 0.57 | 2.27 | |
| GRPOModel Size=Qwen2.5-0.5B2026.06 | 0.56 | 2.25 | |
| GRPOModel Size=Qwen2.5-1.5B2026.06 | 0.28 | 4.56 | |
| BaseModel Size=Qwen2.5-0.5B2026.06 | 0 | 0 | |
| 20-EntropyModel Size=Qwen2.5-0.5B2026.06 | 0 | 0 |