Open-domain Question Answering on MuSiQue Search-o1 framework (test)
20.4AccuracyBest Baseline
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Best BaselineN=2502026.05 | 20.4 | 362 | -56 | |
| PROBE&PREFILLN=2502026.05 | 20.4 | 431 | -48 | |
| DefaultN=2502026.05 | 19.6 | 829 | — |
Loading the SOTA2 catalog…