Long-form Question Answering on Biography
49.7VeriScore F1EWE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| EWEBase Model=Llama-3.1 70B2024.12 | 49.7 | 50.2 | |
| RABase Model=Llama-3.1 70B2024.12 | 43.8 | 49.4 | |
| DRAGINBase Model=Llama-3.1 70B2024.12 | 42.8 | 33.5 | |
| EWEBase Model=Llama-3.1 8B2024.12 | 42.2 | 21.5 | |
| NESTBase Model=Llama-3.1 8B2024.12 | 41.8 | 21.8 | |
| NESTBase Model=Llama-3.1 70B2024.12 | 41.5 | 22.1 | |
| RABase Model=Llama-3.1 8B2024.12 | 41.4 | 21.3 | |
| COVE w/ RetrievalBase Model=Llama-3.1 70B2024.12 | 38.2 | 29.4 | |
| COVEBase Model=Llama-3.1 70B2024.12 | 37.7 | 31.3 | |
| Llama-3.1Base Model=Llama-3.1 70B2024.12 | 37.1 | — | |
| DRAGINBase Model=Llama-3.1 8B2024.12 | 34.7 | 11.4 | |
| COVE w/ RetrievalBase Model=Llama-3.1 8B2024.12 | 29.1 | 10.2 | |
| Llama-3.1Base Model=Llama-3.1 8B2024.12 | 28.9 | 24.2 | |
| COVEBase Model=Llama-3.1 8B2024.12 | 25.1 | 13.3 |