Answer Verification on AIME
0.9648AUROCMSV 64
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MSV 64Sequences=64, Evaluation Protocol=Streaming2026.03 | 0.9648 | 0.0323 | |
| MSV 16Sequences=16, Evaluation Protocol=Streaming2026.03 | 0.9532 | 0.0375 | |
| MSV 4Sequences=4, Evaluation Protocol=Streaming2026.03 | 0.9312 | 0.0472 | |
| ProbeEvaluation Protocol=Streaming (Repeated Rows)2026.03 | 0.912 | 0.0796 | |
| ProbeEvaluation Protocol=Streaming (Row 1)2026.03 | 0.9114 | 0.0796 | |
| MSV 1Sequences=1, Evaluation Protocol=Streaming2026.03 | 0.9056 | 0.0642 | |
| MSV 1 + WV 4Sequences=4, Evaluation Protocol=MSV 1 + Weighted Voting2026.03 | 0.9055 | 0.0657 | |
| MSV 1 + WV 16Sequences=16, Evaluation Protocol=MSV 1 + Weighted Voting2026.03 | 0.9055 | 0.0657 | |
| MSV 1 + WV 64Sequences=64, Evaluation Protocol=MSV 1 + Weighted Voting2026.03 | 0.9055 | 0.0657 | |
| Probe + WV 4Sequences=4, Evaluation Protocol=Weighted Voting2026.03 | 0.874 | 0.0768 | |
| Probe + WV 16Sequences=16, Evaluation Protocol=Weighted Voting2026.03 | 0.874 | 0.0768 | |
| Probe + WV 64Sequences=64, Evaluation Protocol=Weighted Voting2026.03 | 0.874 | 0.0768 |