Multi-task Routing on DISTR-ROUGE2 (test)
26.9ROUGE2SMOOTHIE-LOCAL
Evaluation results
| Method | Links | |
|---|---|---|
| SMOOTHIE-LOCALEnsemble Scale=7B, Neighborhood Size=12024.12 | 26.9 | |
| BEST-MODELEnsemble Scale=7B2024.12 | 26.4 | |
| LABELED-KNNEnsemble Scale=7B, Neighborhood Size=202024.12 | 26.2 | |
| SMOOTHIE-GLOBALEnsemble Scale=7B2024.12 | 26.1 | |
| PAIRRMEnsemble Scale=7B2024.12 | 25.5 | |
| RANDOMEnsemble Scale=7B2024.12 | 25 | |
| SMOOTHIE-LOCALEnsemble Scale=3B, Neighborhood Size=12024.12 | 20.2 | |
| PAIRRMEnsemble Scale=3B2024.12 | 19 | |
| BEST-MODELEnsemble Scale=3B2024.12 | 18.1 | |
| SMOOTHIE-GLOBALEnsemble Scale=3B2024.12 | 18.1 | |
| RANDOMEnsemble Scale=3B2024.12 | 17 | |
| LABELED-KNNEnsemble Scale=3B, Neighborhood Size=202024.12 | 16.8 |