Phrase Grounding On Flickr30K Entities Dev

R@1

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

		Paper Title
Fiber-B	87.1	Coarse-to-Fine Vision-Language Pre-training with Fusion in the Backbone
PEVL	84.1	PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models
VisualBERT	70.4	VisualBERT: A Simple and Performant Baseline for Vision and Language

0 of 3 row(s) selected.