Exploiting Multiple Sequence Lengths in Fast End to End Training for Image Captioning

Open paper
Benchmark Model Rank Results
image-captioning-on-cocoExpansionNet v2#1CIDEr: 143.7
image-captioning-on-coco-captionsExpansionNet v2 (No VL pretraining)#6BLEU-4: 42.7CIDER: 143.7METEOR: 30.6SPICE: 24.7ROUGE-L: 61.1…

Rank counts only results with a code link.