Mathematical Reasoning on AIME (ECE and NLL)
1.65ECEMSV 64
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MSV 64Scale=642026.03 | 1.65 | 0.1254 | |
| MSV 4Scale=42026.03 | 2.71 | 0.1814 | |
| MSV 16Scale=162026.03 | 3.33 | 0.1498 | |
| MSV 64Verifier Type=MSV, Number of Sequences (N)=642026.03 | 3.5 | 0.1501 | |
| MSV 1Scale=12026.03 | 5.11 | 0.2412 | |
| MSV 1 + WV 4Base Scale=1, Weighted Voting=42026.03 | 5.48 | 0.2447 | |
| MSV 1 + WV 16Base Scale=1, Weighted Voting=162026.03 | 5.48 | 0.2447 | |
| MSV 1 + WV 64Base Scale=1, Weighted Voting=642026.03 | 5.48 | 0.2447 | |
| MSV 16Verifier Type=MSV, Number of Sequences (N)=162026.03 | 8.44 | 0.2079 | |
| Probe + WV 4Weighted Voting=42026.03 | 8.61 | 0.2784 | |
| Probe + WV 16Weighted Voting=162026.03 | 8.61 | 0.2784 | |
| Probe + WV 64Weighted Voting=642026.03 | 8.61 | 0.2784 | |
| ProbeContext=Block 12026.03 | 9.01 | 0.2781 | |
| MSV 4Verifier Type=MSV, Number of Sequences (N)=42026.03 | 9.08 | 0.2472 | |
| ProbeVerifier Type=Probe, Number of Sequences (N)=12026.03 | 10.24 | 0.3115 | |
| MSV 1Verifier Type=MSV, Number of Sequences (N)=12026.03 | 10.43 | 0.2993 | |
| ProbeContext=Scaling Blocks2026.03 | 11.6 | 0.2847 | |
| Probe + WV 64Verifier Type=Probe, Number of Sequences (N)=64, Weighted Voting (WV)=true2026.03 | 12.88 | 0.4265 | |
| MSV 1 + WV 64Verifier Type=MSV, Training Sequences=1, Number of Sequences (N)=64, Weighted Voting (WV)=true2026.03 | 13.47 | 0.4444 | |
| Probe + WV 16Verifier Type=Probe, Number of Sequences (N)=16, Weighted Voting (WV)=true2026.03 | 15.79 | 0.5808 | |
| MSV 1 + WV 16Verifier Type=MSV, Training Sequences=1, Number of Sequences (N)=16, Weighted Voting (WV)=true2026.03 | 16.11 | 0.5922 | |
| Probe + WV 4Verifier Type=Probe, Number of Sequences (N)=4, Weighted Voting (WV)=true2026.03 | 27.03 | 1.1423 | |
| MSV 1 + WV 4Verifier Type=MSV, Training Sequences=1, Number of Sequences (N)=4, Weighted Voting (WV)=true2026.03 | 27.14 | 1.1412 |