Long-form Question Answering on LongFact
75.9VeriScore F1EWE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| EWEBase Model=Llama-3.1 70B2024.12 | 75.9 | 50.1 | |
| RABase Model=Llama-3.1 70B2024.12 | 72.7 | 41.2 | |
| DRAGINBase Model=Llama-3.1 70B2024.12 | 71.5 | 38.2 | |
| COVE w/ RetrievalBase Model=Llama-3.1 70B2024.12 | 67.4 | 31.8 | |
| EWEBase Model=Llama-3.1 8B2024.12 | 67.3 | 40.5 | |
| RABase Model=Llama-3.1 8B2024.12 | 65.9 | 28.1 | |
| Llama-3.1Base Model=Llama-3.1 70B2024.12 | 64.3 | — | |
| DRAGINBase Model=Llama-3.1 8B2024.12 | 63.9 | 15.9 | |
| COVEBase Model=Llama-3.1 70B2024.12 | 63.8 | 39.3 | |
| NESTBase Model=Llama-3.1 70B2024.12 | 63.2 | 9.1 | |
| Llama-3.1Base Model=Llama-3.1 8B2024.12 | 63.1 | 40.6 | |
| NESTBase Model=Llama-3.1 8B2024.12 | 62.3 | 4.2 | |
| COVE w/ RetrievalBase Model=Llama-3.1 8B2024.12 | 53.5 | 12.2 | |
| COVEBase Model=Llama-3.1 8B2024.12 | 44.1 | 8.8 |