Scientific Reasoning Question Answering on QASC
74.61AccuracyS-DPO
Evaluation Results
| Method | Links | |
|---|---|---|
| S-DPOModel=Llama32026.05 | 74.61 | |
| DPO-kModel=Llama32026.05 | 73.95 | |
| MASS-DPOModel=Llama32026.05 | 73.62 | |
| MASS-DPOModel=Qwen32026.05 | 72.19 | |
| DPO-kModel=Qwen32026.05 | 71.96 | |
| MASS-DPOModel=SmolLM32026.05 | 71.63 | |
| S-DPOModel=Qwen32026.05 | 71.08 | |
| DPOModel=Llama32026.05 | 71.08 | |
| DPO-kModel=SmolLM32026.05 | 69.98 | |
| DMPOModel=Llama32026.05 | 69.87 | |
| S-DPOModel=SmolLM32026.05 | 69.43 | |
| DPOModel=Qwen32026.05 | 68.43 | |
| DPOModel=SmolLM32026.05 | 67 | |
| DMPOModel=Qwen32026.05 | 66.78 | |
| DMPOModel=SmolLM32026.05 | 65.23 |