Multi-step reasoning and knowledge retrieval on GPQA (test)
52.27AccuracyDynamic Persona Routing
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Dynamic Persona RoutingTotal=448, Sampled=442026.04 | 52.27 | — | 23 | |
| Best Static BaselineTotal=448, Sampled=442026.04 | — | 41.96 | — |