MAIL++: Multi-Modal Bi-directional Agent Layer for Vision-Language Models
Open paper
Benchmark
Model
Rank
Results
prompt-engineering-on-imagenet-a
MAIL++
–
Top-1 accuracy %: 51.10
prompt-engineering-on-imagenet-s
MAIL++
–
Top-1 accuracy %: 50.40
prompt-engineering-on-imagenet-v2
MAIL++
–
Top-1 accuracy %: 65.03
Rank counts only results with a code link.