Multi-hop Question Answering on MuSiQue (EM, F1, Latency)
38.5EMG-reasoner
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| G-reasoner2025.09 | 38.5 | 52.5 | 0.125 | |
| Search-R12025.09 | 37.4 | 53.2 | 0.603 | |
| R1-searcher2025.09 | 34.7 | 48.4 | 0.588 | |
| IRCoT2025.09 | 19.1 | 30.5 | 1.152 |
Loading the SOTA2 catalog…