General Reasoning on GPQA (Acc, Cost)
48.13AccuracyCoT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CoTModel=Gemma32026.03 | 48.13 | 4,926 | |
| CoTModel=Qwen32026.03 | 44.8 | 4,089 | |
| CoTModel=Llama3.12026.03 | 44.06 | 4,156 | |
| DSTModel=Gemma32026.03 | 5.37 | 4,940 | |
| DSTModel=Qwen32026.03 | 4.9 | 4,141 | |
| DSTModel=Llama3.12026.03 | 4.48 | 3,994 | |
| ToTModel=Gemma32026.03 | 4.1 | 10,602 | |
| ToTModel=Qwen32026.03 | 3.76 | 8,875 | |
| ToTModel=Llama3.12026.03 | 3.75 | 8,678 | |
| DPTSModel=Gemma32026.03 | 3.44 | 8,336 | |
| DPTSModel=Qwen32026.03 | 3.27 | 7,001 | |
| DPTSModel=Llama3.12026.03 | 3.14 | 6,892 |