Multi-Hop Question Answering on MuSiQue (Answer F1 Score)
36.3Answer F1IN-CONTEXT
Evaluation Results
| Method | Links | |
|---|---|---|
| IN-CONTEXT2026.02 | 36.3 | |
| SHINETraining Data Scale=Full, Training Data Source=MQA + 1QA2026.02 | 28.5 | |
| SHINETraining Data Scale=1/2, Training Data Source=MQA2026.02 | 28.2 | |
| SHINETraining Data Scale=1/2, Training Data Source=1QA2026.02 | 26.7 | |
| GEN ADAPTER2026.02 | 19.4 | |
| NAIVE2026.02 | 11.8 |