Factual Consistency Evaluation on QAGS-X
0.618Pearson Correlation (rp)BATCHEVAL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| BATCHEVALAPI cost ($/item)=0.05212023.12 | 0.618 | 0.634 | |
| G-Eval*2023.12 | 0.558 | 0.537 | |
| CloserLookAPI cost ($/item)=0.06912023.12 | 0.549 | 0.573 | |
| QAGS*2023.12 | 0.175 | — | |
| BERTScore*2023.12 | 0.024 | 0.008 |