Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Text-to-Image Generation Evaluation on RichHF-18K
Loading...
40.4
Pearson-r
UnifiedReward_Q
4.468
13.7965
23.125
32.4535
Jun 3, 2025
Pearson-r
Kendall's Tau
Updated 1mo ago
Evaluation Results
Method
Method
Links
Pearson-r
Kendall's Tau
UnifiedReward_Q
Scale=8B
2025.06
40.4
-
UnifiedReward_L
Scale=7B
2025.06
39.9
-
Gemini-2.5-Pro
Scale=/
2025.06
39.7
-
Qwen3-VL
Scale=8B
2025.06
38.9
-
Minos
Scale=8B
2025.06
36
-
LLaVA-Critic
Scale=72B
2025.06
33
-
GPT-4o
Scale=/
2025.06
31.1
-
LLaVA-OV
Scale=72B
2025.06
27.2
-
PickScore
2025.06
22.4
18.3
HPS v2
2025.06
19.5
13.1
LLaVA-Critic
Scale=7B
2025.06
18.4
-
Prometheus-V
Scale=7B
2025.06
8.19
-
LLaVA-OV
Scale=7B
2025.06
5.85
-
Feedback
Search any
task
Search any
task