Question Answering on PIQA In-Domain (Accuracy)
89AccuracyFew-shot*
Evaluation Results
| Method | Links | |
|---|---|---|
| Few-shot*Model=Qwen3-32B2025.09 | 89 | |
| Few-shot*Model=Llama3.1-70B2025.09 | 88.6 | |
| ICRModel=Llama3.1-70B2025.09 | 83.2 | |
| ICRModel=Qwen3-32B2025.09 | 82.8 | |
| Zero-shotModel=Qwen3-32B2025.09 | 78 | |
| I2CLModel=Llama3.1-70B2025.09 | 75.4 | |
| I2CLModel=Qwen3-32B2025.09 | 74.2 | |
| FVModel=Llama3.1-70B2025.09 | 71.2 | |
| Zero-shotModel=Llama3.1-70B2025.09 | 70.4 | |
| FVModel=Qwen3-32B2025.09 | 63.4 |