Mathematical Reasoning on AMC 23 (Sel-AUC, AUROC)
0.985Sel-AUCStructU+Self-ConsU
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| StructU+Self-ConsUModel=DeepSeek R1, Aggregation=total2026.06 | 0.985 | 1 | |
| StructUModel=Claude 4.5 Sonnet, Aggregation=within2026.06 | 0.967 | 1 | |
| Self-ConsUModel=Claude 4.5 Sonnet2026.06 | 0.955 | 1 | |
| Self-ConsUModel=Qwen 3 32B2026.06 | 0.81 | 1 | |
| StructU+Self-ConsUModel=Amazon Nova Premier, Aggregation=total2026.06 | 0.718 | 1 |