PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models
Open paper
Benchmark
Model
Rank
Results
phrase-grounding-on-flickr30k-entities-dev
PEVL
#2
R@1: 84.1
phrase-grounding-on-flickr30k-entities-test
PEVL
#4
R@1: 84.4
Rank counts only results with a code link.