Combining inherent knowledge of vision-language models with unsupervised domain adaptation through strong-weak guidance
Open paper
Benchmark
Model
Rank
Results
domain-adaptation-on-domainnet-1
SWG
#3
Accuracy: 66.1
domain-adaptation-on-office-home
SWG
#1
Accuracy: 92.3
domain-adaptation-on-visda2017
SWG
#4
Accuracy: 92.7
Rank counts only results with a code link.