Logical Reasoning on DetectiveQA
88.31Accuracy (DetectiveQA)CDS
Evaluation Results
| Method | Links | |
|---|---|---|
| CDSModel=gpt-5.2, Number of demonstrations=1242026.05 | 88.31 | |
| CDSbgeModel=gpt-5.2, Number of demonstrations=1242026.05 | 88.31 | |
| CDSbgeModel=gpt-5.2, Number of demonstrations=642026.05 | 86.36 | |
| originModel=gpt-5.2, Number of demonstrations=1242026.05 | 85.71 | |
| CDSbgeModel=gpt-5.2, Number of demonstrations=322026.05 | 85.71 | |
| CDSModel=gpt-5.2, Number of demonstrations=642026.05 | 85.06 | |
| originModel=gpt-5.2, Number of demonstrations=642026.05 | 83.77 | |
| CDSModel=gpt-5.2, Number of demonstrations=322026.05 | 83.12 | |
| originModel=gpt-5.2, Number of demonstrations=322026.05 | 82.47 | |
| CDSbgeModel=gpt-5.2, Number of demonstrations=162026.05 | 81.17 | |
| originModel=gpt-5.2, Number of demonstrations=162026.05 | 80.52 | |
| CDSModel=gpt-5.2, Number of demonstrations=162026.05 | 80.52 | |
| CDSbgeModel=Qwen3, Number of demonstrations=322026.05 | 77.29 | |
| CDSbgeModel=Qwen3, Number of demonstrations=1242026.05 | 76.63 | |
| CDSModel=Qwen3, Number of demonstrations=162026.05 | 76.62 | |
| originModel=Qwen3, Number of demonstrations=162026.05 | 75.97 | |
| CDSModel=Qwen3, Number of demonstrations=322026.05 | 75.32 | |
| CDSModel=Qwen3, Number of demonstrations=1242026.05 | 75.32 | |
| CDSbgeModel=Qwen3, Number of demonstrations=162026.05 | 75.32 | |
| CDSbgeModel=Qwen3, Number of demonstrations=642026.05 | 75.32 | |
| originModel=Qwen3, Number of demonstrations=322026.05 | 74.03 | |
| CDSModel=Qwen3, Number of demonstrations=642026.05 | 73.38 | |
| originModel=Qwen3, Number of demonstrations=1242026.05 | 72.73 | |
| originModel=Qwen3, Number of demonstrations=642026.05 | 70.78 |