Multi-hop Question Answering on 2Wiki-Text
58.88EMMyGO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MyGOModel=Llama3.1-8B, Type=MyGO2025.05 | 58.88 | 65.88 | |
| Llama3.1-8BType=WarmUp2025.05 | 58.03 | 64.81 | |
| MyGOModel=Qwen2.5-7B, Type=MyGO2025.05 | 53.17 | 59.62 | |
| Qwen2.5-7BType=WarmUp2025.05 | 50.18 | 56.25 | |
| Llama3.1-8BType=Few-Shot2025.05 | 37.83 | 46.53 | |
| GPT-4.1Type=Few-Shot2025.05 | 30.7 | 52.14 | |
| Qwen2.5-7BType=Few-Shot2025.05 | 23.35 | 33.52 |