Fact Verification on FEVER (F1)
53.9F1 ScoreFLARE
Evaluation Results
| Method | Links | |
|---|---|---|
| FLARELLM=Llama-3.1-8B, Write-back=true2026.03 | 53.9 | |
| FLARELLM=Gemma-3-12B, Write-back=true2026.03 | 51.31 | |
| FLARELLM=Llama-3.1-8B, Write-back=false2026.03 | 48.18 | |
| FLARELLM=Gemma-3-12B, Write-back=false2026.03 | 46.25 | |
| Naive RAGLLM=Llama-3.1-8B, Write-back=true2026.03 | 39.89 | |
| RePlugLLM=Llama-3.1-8B, Write-back=true2026.03 | 39.6 | |
| RePlugLLM=Gemma-3-12B, Write-back=true2026.03 | 37.92 | |
| Naive RAGLLM=Gemma-3-12B, Write-back=true2026.03 | 37.89 | |
| Self-RAGLLM=Llama-3.1-8B, Write-back=true2026.03 | 34.77 | |
| No RetrievalLLM=Gemma-3-12B2026.03 | 34.24 | |
| Naive RAGLLM=Llama-3.1-8B, Write-back=false2026.03 | 34.08 | |
| RePlugLLM=Llama-3.1-8B, Write-back=false2026.03 | 33.96 | |
| No RetrievalLLM=Llama-3.1-8B2026.03 | 33.13 | |
| Naive RAGLLM=Gemma-3-12B, Write-back=false2026.03 | 32.77 | |
| RePlugLLM=Gemma-3-12B, Write-back=false2026.03 | 32.73 | |
| Self-RAGLLM=Gemma-3-12B, Write-back=true2026.03 | 32.08 | |
| Self-RAGLLM=Llama-3.1-8B, Write-back=false2026.03 | 31.45 | |
| Self-RAGLLM=Gemma-3-12B, Write-back=false2026.03 | 29.61 |