Multimodal Preference Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
80.2T2I Accuracy
16
Mar 10, 2026
37.8Tau-b Score
10
Apr 2, 2026
57.1tau_c
10
Apr 2, 2026
57.8tau_c
10
Apr 2, 2026
95.4P-Acc
10
Apr 2, 2026
95.1Preference Accuracy
10
Apr 2, 2026
86.5P-Acc
10
Apr 2, 2026
10.28Kendall's Tau
8
Apr 2, 2026
65.2P-Acc
7
Apr 2, 2026
66.2P-Acc
7
Apr 2, 2026