Factuality and Uncertainty Estimation on Bios (test)
71.4FALOGU-DPO
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| LOGU-DPOBackbone=Llama3-8B-Instruct, Method Category=Training-Based2024.10 | 71.4 | 70.9 | 4.84 | |
| LOGU-DPOBackbone=Mistral-7B-Instruct, Method Category=Training-Based2024.10 | 65.4 | 80.7 | 6.54 | |
| Pair-FewBackbone=Llama3-8B-Instruct, Method Category=Prompt-Based2024.10 | 60.6 | 46.3 | 9.86 | |
| Unc-FewBackbone=Llama3-8B-Instruct, Method Category=Prompt-Based2024.10 | 58.7 | 69.1 | 11.7 | |
| LOGU-SFTBackbone=Llama3-8B-Instruct, Method Category=Training-Based2024.10 | 58.5 | 69.7 | 11.5 | |
| LOGU-SFTBackbone=Mistral-7B-Instruct, Method Category=Training-Based2024.10 | 54.5 | 77.1 | 11.4 | |
| Unc-ZeroBackbone=Llama3-8B-Instruct, Method Category=Prompt-Based2024.10 | 53.8 | 65.4 | 14.9 | |
| Orig.Backbone=Llama3-8B-Instruct, Method Category=Original Model2024.10 | 51.9 | — | 20.4 | |
| Self-RefineBackbone=Llama3-8B-Instruct, Method Category=Prompt-Based2024.10 | 51.8 | 40.5 | 18.8 | |
| Unc-ZeroBackbone=Mistral-7B-Instruct, Method Category=Prompt-Based2024.10 | 43.6 | 76.6 | 22.3 | |
| Unc-FewBackbone=Mistral-7B-Instruct, Method Category=Prompt-Based2024.10 | 41.7 | 74.8 | 22.8 | |
| Pair-FewBackbone=Mistral-7B-Instruct, Method Category=Prompt-Based2024.10 | 40.3 | 66.7 | 25.9 | |
| Orig.Backbone=Mistral-7B-Instruct, Method Category=Original Model2024.10 | 38.8 | — | 27.9 | |
| Self-RefineBackbone=Mistral-7B-Instruct, Method Category=Prompt-Based2024.10 | 38.3 | 57.4 | 27.2 |