Mathematical Reasoning on FrontierScience-Olympiad
50.8AccuracyPC-cubic
Evaluation Results
| Method | Links | |
|---|---|---|
| PC-cubicModel=GPT-OSS-120B, Token Budget (B)=1M2026.05 | 50.8 | |
| PC-cubicModel=GPT-OSS-120B, Token Budget (B)=5M2026.05 | 50.7 | |
| PC-cubicModel=GPT-OSS-120B, Token Budget (B)=250k2026.05 | 50.6 | |
| PC-quadraticModel=GPT-OSS-120B, Token Budget (B)=1M2026.05 | 50.4 | |
| PC-quadraticModel=Nemotron3-30B, Token Budget (B)=5M2026.05 | 50.4 | |
| PC-quadraticModel=GPT-OSS-120B, Token Budget (B)=5M2026.05 | 50.3 | |
| PC-cubicModel=Nemotron3-30B, Token Budget (B)=5M2026.05 | 50.3 | |
| PC-quadraticModel=GPT-OSS-120B, Token Budget (B)=250k2026.05 | 50.2 | |
| P(True)Model=GPT-OSS-120B, Token Budget (B)=1M2026.05 | 50 | |
| P(True)Model=GPT-OSS-120B, Token Budget (B)=5M2026.05 | 50 | |
| Standard MVModel=GPT-OSS-120B, Token Budget (B)=5M2026.05 | 49.9 | |
| Self-certaintyModel=GPT-OSS-120B, Token Budget (B)=5M2026.05 | 49.7 | |
| DeepConf tailModel=GPT-OSS-120B, Token Budget (B)=5M2026.05 | 49.7 | |
| PC-linearModel=GPT-OSS-120B, Token Budget (B)=1M2026.05 | 49.6 | |
| PC-linearModel=GPT-OSS-120B, Token Budget (B)=5M2026.05 | 49.6 | |
| Standard MVModel=GPT-OSS-120B, Token Budget (B)=1M2026.05 | 49.5 | |
| Self-certaintyModel=GPT-OSS-120B, Token Budget (B)=1M2026.05 | 49.5 | |
| DeepConf tailModel=GPT-OSS-120B, Token Budget (B)=1M2026.05 | 49.5 | |
| PC-linearModel=GPT-OSS-120B, Token Budget (B)=250k2026.05 | 49.5 | |
| Self-certaintyModel=GPT-OSS-120B, Token Budget (B)=250k2026.05 | 49.4 | |
| DeepConf tailModel=Nemotron3-30B, Token Budget (B)=5M2026.05 | 49.4 | |
| PC-linearModel=Nemotron3-30B, Token Budget (B)=5M2026.05 | 49.4 | |
| Standard MVModel=GPT-OSS-120B, Token Budget (B)=250k2026.05 | 49.3 | |
| DeepConf tailModel=GPT-OSS-120B, Token Budget (B)=250k2026.05 | 49.3 | |
| P(True)Model=GPT-OSS-120B, Token Budget (B)=250k2026.05 | 49.3 | |
| P(True)Model=Nemotron3-30B, Token Budget (B)=5M2026.05 | 49.2 | |
| Self-certaintyModel=Nemotron3-30B, Token Budget (B)=5M2026.05 | 49.1 | |
| Standard MVModel=Nemotron3-30B, Token Budget (B)=5M2026.05 | 49 | |
| PC-quadraticModel=Nemotron3-30B, Token Budget (B)=1M2026.05 | 48.6 | |
| PC-cubicModel=Nemotron3-30B, Token Budget (B)=1M2026.05 | 48.6 | |
| PC-linearModel=Nemotron3-30B, Token Budget (B)=1M2026.05 | 48.1 | |
| DeepConf tailModel=Nemotron3-30B, Token Budget (B)=1M2026.05 | 47 | |
| Standard MVModel=Nemotron3-30B, Token Budget (B)=1M2026.05 | 46.8 | |
| Self-certaintyModel=Nemotron3-30B, Token Budget (B)=1M2026.05 | 46.8 | |
| P(True)Model=Nemotron3-30B, Token Budget (B)=1M2026.05 | 46.5 | |
| PC-quadraticModel=Nemotron3-30B, Token Budget (B)=250k2026.05 | 43.3 | |
| PC-cubicModel=Nemotron3-30B, Token Budget (B)=250k2026.05 | 43.3 | |
| PC-linearModel=Nemotron3-30B, Token Budget (B)=250k2026.05 | 42.9 | |
| DeepConf tailModel=Nemotron3-30B, Token Budget (B)=250k2026.05 | 40.5 | |
| Standard MVModel=Nemotron3-30B, Token Budget (B)=250k2026.05 | 40.4 | |
| Self-certaintyModel=Nemotron3-30B, Token Budget (B)=250k2026.05 | 40.2 | |
| P(True)Model=Nemotron3-30B, Token Budget (B)=250k2026.05 | 39.4 | |
| PC-cubicModel=Ministral3-14B, Token Budget (B)=5M2026.05 | 18.1 | |
| PC-cubicModel=Ministral3-14B, Token Budget (B)=1M2026.05 | 17.4 | |
| PC-quadraticModel=Ministral3-14B, Token Budget (B)=5M2026.05 | 17.3 | |
| PC-quadraticModel=Ministral3-14B, Token Budget (B)=1M2026.05 | 17 | |
| PC-linearModel=Ministral3-14B, Token Budget (B)=5M2026.05 | 16.2 | |
| PC-linearModel=Ministral3-14B, Token Budget (B)=1M2026.05 | 16 | |
| PC-cubicModel=Ministral3-14B, Token Budget (B)=250k2026.05 | 15.8 | |
| PC-quadraticModel=Ministral3-14B, Token Budget (B)=250k2026.05 | 15.6 | |
| PC-linearModel=Ministral3-14B, Token Budget (B)=250k2026.05 | 14.9 | |
| P(True)Model=Ministral3-14B, Token Budget (B)=5M2026.05 | 14.6 | |
| P(True)Model=Ministral3-14B, Token Budget (B)=1M2026.05 | 14.3 | |
| Standard MVModel=Ministral3-14B, Token Budget (B)=5M2026.05 | 14.2 | |
| Standard MVModel=Ministral3-14B, Token Budget (B)=1M2026.05 | 14.1 | |
| Standard MVModel=Ministral3-14B, Token Budget (B)=250k2026.05 | 13.7 | |
| Self-certaintyModel=Ministral3-14B, Token Budget (B)=5M2026.05 | 13.7 | |
| DeepConf tailModel=Ministral3-14B, Token Budget (B)=5M2026.05 | 13.7 | |
| P(True)Model=Ministral3-14B, Token Budget (B)=250k2026.05 | 13.7 | |
| Self-certaintyModel=Ministral3-14B, Token Budget (B)=1M2026.05 | 13.5 | |
| DeepConf tailModel=Ministral3-14B, Token Budget (B)=1M2026.05 | 13.4 | |
| Self-certaintyModel=Ministral3-14B, Token Budget (B)=250k2026.05 | 13 | |
| DeepConf tailModel=Ministral3-14B, Token Budget (B)=250k2026.05 | 13 |