Strategic Reasoning on CheckmateInOne
65.33Acc@t1SELF-THOUGHT
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SELF-THOUGHTModel=GPT-4O2026.01 | 65.33 | 32 | 8 | |
| INITIAL (t = 0)Model=GPT-4O2026.01 | 41.33 | — | — | |
| SELF-REFINEModel=GPT-4O2026.01 | 38.67 | 13.33 | 16 | |
| SELF-THOUGHTModel=O3-MINI2026.01 | 37.33 | 21.33 | 18.67 | |
| INITIAL (t = 0)Model=O3-MINI2026.01 | 34.67 | — | — | |
| SELF-THOUGHTModel=GPT-4O-MINI2026.01 | 33.33 | 14.67 | 12 | |
| REFLEXModel=O3-MINI2026.01 | 32 | 5.33 | 8 | |
| REFLEXIONModel=O3-MINI2026.01 | 32 | 12 | 14.67 | |
| INITIAL (t = 0)Model=GPT-4O-MINI2026.01 | 30.67 | — | — | |
| SELF-TICKModel=GPT-4O2026.01 | 30.67 | 16 | 26.67 | |
| REFLEXModel=GPT-4O2026.01 | 26.67 | 12 | 26.67 | |
| REFLEXIONModel=GPT-4O2026.01 | 25.33 | 16 | 32 | |
| SELF-TICKModel=GPT-4O-MINI2026.01 | 20 | 6.67 | 17.33 | |
| SELF-REFINEModel=O3-MINI2026.01 | 20 | 4 | 18.67 | |
| SELF-THOUGHTModel=DEEPSEEK-R12026.01 | 20 | 12 | 9.33 | |
| REFLEXIONModel=DEEPSEEK-R12026.01 | 18.67 | 14.67 | 13.33 | |
| INITIAL (t = 0)Model=DEEPSEEK-R12026.01 | 17.33 | — | — | |
| REFLEXModel=DEEPSEEK-R12026.01 | 16 | 9.33 | 10.67 | |
| SELF-REFINEModel=DEEPSEEK-R12026.01 | 16 | 10.67 | 12 | |
| SELF-TICKModel=O3-MINI2026.01 | 13.33 | 2.67 | 24 | |
| SELF-REFINEModel=GPT-4O-MINI2026.01 | 10.67 | 6.67 | 26.67 | |
| REFLEXModel=GPT-4O-MINI2026.01 | 9.33 | 5.33 | 26.67 | |
| REFLEXIONModel=GPT-4O-MINI2026.01 | 9.33 | 4 | 25.33 | |
| SELF-TICKModel=DEEPSEEK-R12026.01 | 5.33 | 1.33 | 13.33 |