Defective Dialog Detection on Multi-domain n = 714 (test)
84PrecisionMean TLD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Mean TLDaggregation_strategy=mean2023.06 | 84 | 54 | 66 | |
| Last-turn TLDaggregation_strategy=last-turn2023.06 | 83 | 68 | 75 | |
| Rising linear weightsaggregation_strategy=linear weighting2023.06 | 83 | 63 | 72 | |
| Union of mean & last-turn TLDaggregation_strategy=union2023.06 | 82 | 73 | 77 | |
| DQMmodel_type=supervised model2023.06 | 78 | 83 | 81 |