Question Answering on TruthfulQA (PRR)
53.5PRRTAD
Evaluation Results
| Method | Links | |
|---|---|---|
| TADBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 53.5 | |
| MINDBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 49 | |
| SAPLMABackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 44.9 | |
| Semantic DensityBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 44.8 | |
| LookBackLensBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 43.1 | |
| SheepsBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 40.9 | |
| LUQBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 35.2 | |
| SATRMDBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 31.5 | |
| Mean Token EntropyBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 31 | |
| FocusBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 30.6 | |
| PerplexityBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 27.8 | |
| Simple FocusBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 27 | |
| SARBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 26 | |
| EVL NLI Score entail.Backbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 23.6 | |
| Ecc. NLI Score entail.Backbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 23 | |
| DegMat NLI Score entail.Backbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 20.6 | |
| Lexical Similarity Rouge-LBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 13.7 | |
| MSPBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 12.6 | |
| EigenScoreBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 10.6 | |
| FactoscopeBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 7.7 | |
| CCPBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 4 | |
| Semantic EntropyBackbone=Gemma-2 9b, Reference Quality Metric=AlignScore2024.08 | 0.7 |