Natural Language Inference on CB Far OOD
94.6Accuracy (CB Far OOD)ICR
Evaluation Results
| Method | Links | |
|---|---|---|
| ICRModel=Qwen3-32B2025.09 | 94.6 | |
| ICRModel=Llama3.1-70B2025.09 | 93.4 | |
| Few-shot*Model=Llama3.1-70B2025.09 | 92.9 | |
| Zero-shotModel=Qwen3-32B2025.09 | 91.1 | |
| Few-shot*Model=Qwen3-32B2025.09 | 91.1 | |
| Zero-shotModel=Llama3.1-70B2025.09 | 91 | |
| I2CLModel=Llama3.1-70B2025.09 | 90.2 | |
| I2CLModel=Qwen3-32B2025.09 | 89.6 | |
| FVModel=Llama3.1-70B2025.09 | 85.6 | |
| FVModel=Qwen3-32B2025.09 | 85.2 |