Selective Prediction on MATH 500
95Sel-AUCStructU+Self-ConsU (TrueSkill+PageRank)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| StructU+Self-ConsU (TrueSkill+PageRank)Model=Claude 4.5 Sonnet, Scope=across2026.06 | 95 | 83.4 | |
| StructU+Self-ConsU (TrueSkill+PageRank)Model=DeepSeek R1, Scope=across2026.06 | 94.2 | 80 | |
| StructU+Self-ConsU (TrueSkill+PageRank)Model=GPT-OSS 20B, Scope=total2026.06 | 90.5 | 72.6 | |
| StructU+Self-ConsU (TrueSkill+PageRank)Model=Qwen 3 32B, Scope=total2026.06 | 88.8 | 86.7 | |
| StructU+Self-ConsU (TrueSkill+PageRank)Model=Amazon Nova Premier, Scope=total2026.06 | 88.5 | 86.5 |