Text Summarization on Newsroom segment-level
0.684Coherence (COH)GPTScore
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| GPTScore2026.05 | 0.684 | 0.705 | 0.662 | 0.614 | |
| DIFFSCORE-FTFine-tuned=true2026.05 | 0.683 | 0.7 | 0.658 | 0.618 | |
| DIFFSCORE2026.05 | 0.675 | 0.676 | 0.651 | 0.625 | |
| BARTScore2026.05 | 0.666 | 0.653 | 0.63 | 0.584 | |
| BLEU2026.05 | 0.578 | 0.484 | 0.733 | 0.636 | |
| QuestEval2026.05 | 0.569 | 0.486 | 0.649 | 0.561 | |
| METEOR2026.05 | 0.565 | 0.488 | 0.715 | 0.623 | |
| UniEval2026.05 | 0.51 | 0.563 | 0.407 | 0.371 | |
| AlignScore2026.05 | 0.489 | 0.486 | 0.478 | 0.5 | |
| G-Eval2026.05 | 0.409 | 0.484 | 0.504 | 0.536 | |
| MoverScore2026.05 | 0.165 | 0.141 | 0.205 | 0.216 | |
| BERTScore2026.05 | 0.133 | 0.133 | 0.111 | 0.148 | |
| ROUGE-12026.05 | 0.084 | 0.091 | 0.114 | 0.131 | |
| ROUGE-22026.05 | 0.027 | 0.048 | 0.08 | 0.093 | |
| ROUGE-L2026.05 | 0.027 | 0.045 | 0.044 | 0.077 |