Unified Vision-Language Pre-Training for Image Captioning and VQA
Open paper
Benchmark
Model
Rank
Results
image-captioning-on-flickr30k-captions-test
Unified VLP
#1
BLEU-4: 30.1
CIDEr: 67.4
METEOR: 23
SPICE: 17
visual-question-answering-on-vqa-v2-test-std
Unified VLP
#27
overall: 70.7
Rank counts only results with a code link.