Uncertainty Quantification on Aggregate All 11 Datasets
55.7Mean PRRTAD
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TADBackbone=Gemma-2 9b2024.08 | 55.7 | 2.36 | |
| TADBackbone=Llama-3.1 8b2024.08 | 54.5 | 1.82 | |
| LookBackLensBackbone=Gemma-2 9b2024.08 | 53 | 4.64 | |
| SheepsBackbone=Llama-3.1 8b2024.08 | 51 | 3.09 | |
| LookBackLensBackbone=Llama-3.1 8b2024.08 | 49.9 | 4.45 | |
| SATRMDBackbone=Gemma-2 9b2024.08 | 48.4 | 4 | |
| SATRMDBackbone=Llama-3.1 8b2024.08 | 48.1 | 4.45 | |
| MINDBackbone=Gemma-2 9b2024.08 | 46.4 | 7.55 | |
| SheepsBackbone=Gemma-2 9b2024.08 | 45.7 | 8.09 | |
| MINDBackbone=Llama-3.1 8b2024.08 | 45 | 6.36 | |
| SAPLMABackbone=Gemma-2 9b2024.08 | 40.2 | 9.36 | |
| FocusBackbone=Gemma-2 9b2024.08 | 38.4 | 10.18 | |
| Simple FocusBackbone=Gemma-2 9b2024.08 | 36.6 | 9.45 | |
| Mean Token EntropyBackbone=Gemma-2 9b2024.08 | 35.6 | 10.45 | |
| PerplexityBackbone=Gemma-2 9b2024.08 | 34.8 | 10.91 | |
| MSPBackbone=Gemma-2 9b2024.08 | 34.3 | 10.27 | |
| MSPBackbone=Llama-3.1 8b2024.08 | 33.6 | 10.55 | |
| Simple FocusBackbone=Llama-3.1 8b2024.08 | 33.5 | 10.45 | |
| PerplexityBackbone=Llama-3.1 8b2024.08 | 33.4 | 11.36 | |
| CCPBackbone=Gemma-2 9b2024.08 | 32 | 12.64 | |
| Semantic DensityBackbone=Gemma-2 9b2024.08 | 32 | 11.64 | |
| Mean Token EntropyBackbone=Llama-3.1 8b2024.08 | 31.3 | 11.55 | |
| SARBackbone=Gemma-2 9b2024.08 | 31.1 | 11.73 | |
| SAPLMABackbone=Llama-3.1 8b2024.08 | 30.8 | 11.82 | |
| FocusBackbone=Llama-3.1 8b2024.08 | 30.5 | 12.55 | |
| LUQBackbone=Gemma-2 9b2024.08 | 30.2 | 13.82 | |
| CCPBackbone=Llama-3.1 8b2024.08 | 28.7 | 13.64 | |
| DegMat NLI Score entail.Backbone=Gemma-2 9b2024.08 | 28.3 | 13.91 | |
| EVL NLI Score entail.Backbone=Gemma-2 9b2024.08 | 27.9 | 14.91 | |
| Semantic DensityBackbone=Llama-3.1 8b2024.08 | 27.3 | 11.82 | |
| SARBackbone=Llama-3.1 8b2024.08 | 25.8 | 12.45 | |
| Lexical Similarity Rouge-LBackbone=Gemma-2 9b2024.08 | 25.4 | 14.82 | |
| LUQBackbone=Llama-3.1 8b2024.08 | 24.6 | 13.55 | |
| DegMat NLI Score entail.Backbone=Llama-3.1 8b2024.08 | 24.4 | 13 | |
| Semantic EntropyBackbone=Gemma-2 9b2024.08 | 23.7 | 16.64 | |
| Ecc. NLI Score entail.Backbone=Gemma-2 9b2024.08 | 23.6 | 17.45 | |
| EVL NLI Score entail.Backbone=Llama-3.1 8b2024.08 | 23.2 | 14.55 | |
| FactoscopeBackbone=Llama-3.1 8b2024.08 | 22.2 | 17.45 | |
| Lexical Similarity Rouge-LBackbone=Llama-3.1 8b2024.08 | 21.6 | 16.18 | |
| Semantic EntropyBackbone=Llama-3.1 8b2024.08 | 21.2 | 16.91 | |
| Ecc. NLI Score entail.Backbone=Llama-3.1 8b2024.08 | 20.5 | 16.82 | |
| EigenScoreBackbone=Gemma-2 9b2024.08 | 20 | 19.18 | |
| EigenScoreBackbone=Llama-3.1 8b2024.08 | 19.1 | 18.18 | |
| FactoscopeBackbone=Gemma-2 9b2024.08 | 15.1 | 19 |