Mathematical Reasoning on AMC (ACC, Length)
74.6AccuracyTACLer
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TACLerMode=Thinking2026.01 | 74.6 | 4,871 | |
| FastCuRLMode=Thinking2026.01 | 73.9 | 6,756 | |
| DeepScaleRMode=Thinking2026.01 | 73.8 | 5,616 | |
| TACLerMode=NoThinking2026.01 | 72.7 | 4,312 | |
| AdaptThinkMode=Efficient Reasoning2026.01 | 67.4 | 5,489 | |
| AutoThinkMode=Efficient Reasoning2026.01 | 66.7 | 4,596 | |
| STILL-3Mode=Thinking2026.01 | 66.3 | 7,091 | |
| R1-QwenMode=Thinking2026.01 | 60.8 | 8,754 | |
| R1-QwenMode=Thinking2026.01 | 60.8 | 8,754 | |
| R1-QwenMode=NoThinking2026.01 | 48.6 | 2,264 |