Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Preference Evaluation on FOILR1
Loading...
95.1
Preference Accuracy
BLIP-S
86.884
89.017
91.15
93.283
Oct 1, 2025
Preference Accuracy
Updated 17d ago
Evaluation Results
Method
Method
Links
Preference Accuracy
BLIP-S
reference-based=false
2025.10
95.1
VL-GUIDE-S-VLM
reference-based=false
2025.10
95
ImgREW-S
reference-based=false
2025.10
93.8
PAC-S
reference-based=false
2025.10
93.7
Polos
reference-based=true
2025.10
93.3
LongCLIP-S
reference-based=false
2025.10
91.6
RefCLIP-S
reference-based=true
2025.10
91
Ref-free Polos
reference-based=false
2025.10
88.7
RefPAC-S
reference-based=true
2025.10
88.7
CLIP-S
reference-based=false
2025.10
87.2
Feedback
Search any
task
Search any
task