Long-form Question Answering on AlpacaFact
66.9VeriScore F1EWE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| EWEBase Model=Llama-3.1 70B2024.12 | 66.9 | 49.9 | |
| RABase Model=Llama-3.1 70B2024.12 | 66 | 43.1 | |
| EWEBase Model=Llama-3.1 8B2024.12 | 65.5 | 28 | |
| DRAGINBase Model=Llama-3.1 70B2024.12 | 65.3 | 31.5 | |
| Llama-3.1Base Model=Llama-3.1 8B2024.12 | 65.3 | 26.7 | |
| COVE w/ RetrievalBase Model=Llama-3.1 70B2024.12 | 64 | 28.8 | |
| RABase Model=Llama-3.1 8B2024.12 | 63.9 | 18.5 | |
| Llama-3.1Base Model=Llama-3.1 70B2024.12 | 63.8 | — | |
| COVEBase Model=Llama-3.1 70B2024.12 | 61.5 | 33.3 | |
| DRAGINBase Model=Llama-3.1 8B2024.12 | 61.3 | 11.1 | |
| NESTBase Model=Llama-3.1 70B2024.12 | 58.1 | 30.2 | |
| NESTBase Model=Llama-3.1 8B2024.12 | 57.8 | 9.1 | |
| COVE w/ RetrievalBase Model=Llama-3.1 8B2024.12 | 54.6 | 12.5 | |
| COVEBase Model=Llama-3.1 8B2024.12 | 51.3 | 15.1 |