Human Judgment Correlation On Flickr8K Cf
평가 지표
Kendall's Tau-b
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | Kendall's Tau-b | Paper Title | Repository |
---|---|---|---|
CLIP-S | 34.4 | CLIPScore: A Reference-free Evaluation Metric for Image Captioning | |
MID | 37.3 | Mutual Information Divergence: A Unified Metric for Multimodal Generative Models | |
RefCLIP-S | 36.4 | CLIPScore: A Reference-free Evaluation Metric for Image Captioning |
0 of 3 row(s) selected.