Fine-grained Image Captioning with CLIP Reward
Open paper
Benchmark
Model
Rank
Results
image-captioning-on-coco-captions
CLIP Text Encoder (RL w/ CIDEr-reward)
#25
BLEU-4: 38.2
CIDER: 124.9
METEOR: 28.7
ROUGE-L: 58.5
Rank counts only results with a code link.