ManagerTower: Aggregating the Insights of Uni-Modal Experts for Vision-Language Representation Learning
37690Knowledge Triples
| Source Entity | Relationship | Target Entity |
|---|---|---|
|
ManagerTower
Method |
achieves |
Flickr30k
Material |
|
ManagerTower
Method |
achieves |
VQAv2 Test Std
Task |
|
ManagerTower
Method |
measures |
Accuracy
Metric |
|
ManagerTower
Method |
measures |
IR@1
Metric |
|
ManagerTower
Method |
measures |
TR@1
Metric |