Selective Prediction on AIME 24
92.5Sel-AUCStructU+Self-ConsU (TrueSkill+PageRank)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| StructU+Self-ConsU (TrueSkill+PageRank)Model=DeepSeek R1, Scope=total2026.06 | 92.5 | 100 | |
| StructU+Self-ConsU (TrueSkill+PageRank)Model=GPT-OSS 20B, Scope=total2026.06 | 89.7 | 100 | |
| StructU+Self-ConsU (TrueSkill+PageRank)Model=Claude 4.5 Sonnet, Scope=total2026.06 | 68.2 | 100 | |
| StructU+Self-ConsU (TrueSkill+PageRank)Model=Qwen 3 32B, Scope=total2026.06 | 42.6 | — | |
| SemanticUModel=Amazon Nova Premier2026.06 | 29.4 | — | |
| StructU (TrueSkill+PageRank)Model=Amazon Nova Premier, Scope=within2026.06 | 23.3 | — |