Factuality Hallucination Evaluation on BioGEN (test)
50.5FactScoreNOVA
Evaluation Results
| Method | Links | |
|---|---|---|
| NOVAInstruction Dataset=Alpaca - GPT4, Sample Ratio=5%2025.02 | 50.5 | |
| NOVAInstruction Dataset=Alpaca, Sample Ratio=5%2025.02 | 50.3 | |
| SELF-EVALInstruction Dataset=Alpaca, Sample Ratio=100%2025.02 | 48.3 | |
| NuggetsInstruction Dataset=Alpaca, Sample Ratio=5%2025.02 | 48.2 | |
| IFDInstruction Dataset=Alpaca, Sample Ratio=5%2025.02 | 48.1 | |
| NOVAInstruction Dataset=Alpaca - GPT4, Sample Ratio=10%2025.02 | 48.1 | |
| CaRInstruction Dataset=Alpaca, Sample Ratio=5%2025.02 | 47.9 | |
| FLAME-DPO factInstruction Dataset=Alpaca, Sample Ratio=100%2025.02 | 47.2 | |
| SELF-EVALInstruction Dataset=Alpaca - GPT4, Sample Ratio=100%2025.02 | 47.2 | |
| NuggetsInstruction Dataset=Alpaca - GPT4, Sample Ratio=5%2025.02 | 47.2 | |
| CaRInstruction Dataset=Alpaca - GPT4, Sample Ratio=5%2025.02 | 46.9 | |
| NOVAInstruction Dataset=Alpaca - GPT4, Sample Ratio=15%2025.02 | 46.9 | |
| NOVAInstruction Dataset=Alpaca, Sample Ratio=10%2025.02 | 46.8 | |
| NuggetsInstruction Dataset=Alpaca - GPT4, Sample Ratio=10%2025.02 | 46.8 | |
| IFDInstruction Dataset=Alpaca - GPT4, Sample Ratio=5%2025.02 | 46.7 | |
| FLAME-DPO factInstruction Dataset=Alpaca - GPT4, Sample Ratio=100%2025.02 | 46.3 | |
| NOVAInstruction Dataset=Alpaca, Sample Ratio=15%2025.02 | 45.9 | |
| CaRInstruction Dataset=Alpaca - GPT4, Sample Ratio=10%2025.02 | 45.9 | |
| NuggetsInstruction Dataset=Alpaca, Sample Ratio=10%2025.02 | 45.8 | |
| CaRInstruction Dataset=Alpaca, Sample Ratio=10%2025.02 | 45.2 | |
| NuggetsInstruction Dataset=Alpaca - GPT4, Sample Ratio=15%2025.02 | 44.8 | |
| CaRInstruction Dataset=Alpaca - GPT4, Sample Ratio=15%2025.02 | 44.6 | |
| NuggetsInstruction Dataset=Alpaca, Sample Ratio=15%2025.02 | 44.3 | |
| CaRInstruction Dataset=Alpaca, Sample Ratio=15%2025.02 | 43.9 | |
| IFDInstruction Dataset=Alpaca - GPT4, Sample Ratio=10%2025.02 | 43.6 | |
| IFDInstruction Dataset=Alpaca, Sample Ratio=10%2025.02 | 43.2 | |
| IFDInstruction Dataset=Alpaca - GPT4, Sample Ratio=15%2025.02 | 42.9 | |
| VanillaInstruction Dataset=Alpaca, Sample Ratio=100%2025.02 | 42.4 | |
| IFDInstruction Dataset=Alpaca, Sample Ratio=15%2025.02 | 42.2 | |
| VanillaInstruction Dataset=Alpaca - GPT4, Sample Ratio=100%2025.02 | 41.9 |