Hallucination Detection on MLQE-PE RO-EN (test)
99.72AUROCLaBSE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LaBSEType=External Detector2022.12 | 99.72 | 49 | |
| CometKiwiType=External Detector, Note=Trained on MLQE-PE data samples2022.12 | 99.62 | 49 | |
| Wass-to-UnifType=Model-based Detector2022.12 | 99.3 | 46 | |
| Attn-ign-SRCType=Model-based Detector2022.12 | 99.16 | 93 | |
| Wass-ComboType=Model-based Detector, Note=Mean and SD across five random seeds2022.12 | 98.75 | 46 | |
| Wass-to-DataType=Model-based Detector, Note=Mean and SD across five random seeds2022.12 | 96.54 | 1,036 | |
| Seq-LogprobType=Model-based Detector2022.12 | 91.97 | 1,642 |