Multi-hop Question Answering on Hotpot Kimi
54.07EMMyGO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MyGOType=MyGO, Backbone=Qwen2.5-7B2025.05 | 54.07 | 68.48 | |
| Qwen2.5-7BType=WarmUp, Backbone=Qwen2.5-7B2025.05 | 52.51 | 66.04 | |
| GPT-4.1Type=Few-Shot, Backbone=GPT-4.12025.05 | 33.5 | 53.35 | |
| Qwen2.5-7BType=Few-Shot, Backbone=Qwen2.5-7B2025.05 | 15.6 | 32.7 |