Mathematical Reasoning on AIME 25 (Accuracy, Pass@4)
58.33AccuracyASLEC-CASL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ASLEC-CASLBase Model=Qwen 4B-Instruct2026.04 | 58.33 | 83.33 | |
| ASLEC-DROPBase Model=Qwen 4B-Instruct2026.04 | 56.66 | 80 | |
| GRACEBase Model=Qwen 4B-Instruct2026.04 | 50 | 73.33 | |
| Local LPBase Model=Qwen 4B-Instruct2026.04 | 49.16 | 73.33 | |
| ASLEC-CASLBase Model=Qwen 8B-Base2026.04 | 37.5 | 53.33 | |
| ASLEC-DROPBase Model=Qwen 8B-Base2026.04 | 36.66 | 43.33 | |
| ASLEC-CASLBase Model=Qwen 4B-Base2026.04 | 30.83 | 46.66 | |
| ASLEC-DROPBase Model=Qwen 4B-Base2026.04 | 28.33 | 43.33 | |
| ASLEC-CASLBase Model=Qwen 7B-Instruct2026.04 | 24.16 | 46.66 | |
| GRACEBase Model=Qwen 8B-Base2026.04 | 21.66 | 36.66 | |
| Local LPBase Model=Qwen 4B-Base2026.04 | 20.83 | 36.66 | |
| Local LPBase Model=Qwen 8B-Base2026.04 | 20.83 | 36.66 | |
| ASLEC-DROPBase Model=Qwen 7B-Instruct2026.04 | 20.83 | 36.66 | |
| GRACEBase Model=Qwen 4B-Base2026.04 | 15.83 | 33.33 | |
| GRACEBase Model=Qwen 7B-Instruct2026.04 | 11.66 | 23.33 | |
| Local LPBase Model=Qwen 7B-Instruct2026.04 | 10.83 | 26.66 |