Mutual Information Divergence: A Unified Metric for Multimodal Generative Models
Open paper
Benchmark
Model
Rank
Results
human-judgment-classification-on-pascal-50s
MID
#1
Mean Accuracy: 85.2
human-judgment-correlation-on-flickr8k-cf
MID
#1
Kendall's Tau-b: 37.3
human-judgment-correlation-on-flickr8k-expert
MID
#1
Kendall's Tau-c: 54.9
Rank counts only results with a code link.