Competition Mathematics on AIME 2024 (Acc@t1 and Delta Analysis)
86.67Acc@t1SELF-THOUGHT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SELF-THOUGHTModel=O3-MINI2026.01 | 86.67 | 6.67 | 0 | |
| SELF-REFINEModel=O3-MINI2026.01 | 83.33 | 10 | 6.67 | |
| INITIAL (t = 0)Model=O3-MINI2026.01 | 80 | — | — | |
| REFLEXModel=O3-MINI2026.01 | 80 | 3.33 | 3.33 | |
| REFLEXIONModel=O3-MINI2026.01 | 80 | 3.33 | 3.33 | |
| INITIAL (t = 0)Model=DEEPSEEK-R12026.01 | 80 | — | — | |
| SELF-THOUGHTModel=DEEPSEEK-R12026.01 | 80 | 10 | 10 | |
| SELF-TICKModel=O3-MINI2026.01 | 76.67 | 3.33 | 6.67 | |
| REFLEXModel=DEEPSEEK-R12026.01 | 76.67 | 0 | 3.33 | |
| SELF-REFINEModel=DEEPSEEK-R12026.01 | 76.67 | 3.33 | 6.67 | |
| SELF-TICKModel=DEEPSEEK-R12026.01 | 60 | 0 | 20 | |
| REFLEXIONModel=DEEPSEEK-R12026.01 | 56.67 | 3.33 | 26.67 | |
| SELF-THOUGHTModel=GPT-4O2026.01 | 33.33 | 20 | 0 | |
| SELF-THOUGHTModel=GPT-4O-MINI2026.01 | 30 | 16.67 | 6.67 | |
| SELF-TICKModel=GPT-4O-MINI2026.01 | 23.33 | 6.67 | 3.33 | |
| INITIAL (t = 0)Model=GPT-4O-MINI2026.01 | 20 | — | — | |
| SELF-REFINEModel=GPT-4O2026.01 | 20 | 10 | 3.33 | |
| SELF-TICKModel=GPT-4O2026.01 | 16.67 | 3.33 | 0 | |
| REFLEXIONModel=GPT-4O2026.01 | 16.67 | 10 | 6.67 | |
| SELF-REFINEModel=GPT-4O-MINI2026.01 | 13.33 | 3.33 | 10 | |
| REFLEXIONModel=GPT-4O-MINI2026.01 | 13.33 | 6.67 | 13.33 | |
| INITIAL (t = 0)Model=GPT-4O2026.01 | 13.33 | — | — | |
| REFLEXModel=GPT-4O2026.01 | 13.33 | 0 | 0 | |
| REFLEXModel=GPT-4O-MINI2026.01 | 10 | 0 | 10 |