Human Judgment Correlation On Flickr8K Expert
Metrics
Kendall's Tau-c
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | Kendall's Tau-c |
|---|---|
| clipscore-a-reference-free-evaluation-metric | 51.2 |
| factual-a-benchmark-for-faithful-and | 54.2 |
| mutual-information-divergence-a-unified | 54.9 |
| clipscore-a-reference-free-evaluation-metric | 53.0 |