NLP-685

Reported on 8 benchmarks across 1 task

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing8 results

Image CaptioningonVizWiz 2020 test
B1
66.62
best: 72.77 (IBM Research AI)
Image CaptioningonVizWiz 2020 test
B2
47.41
best: 54.17 (IBM Research AI)
Image CaptioningonVizWiz 2020 test
B3
32.66
best: 38.97 (IBM Research AI)
Image CaptioningonVizWiz 2020 test
B4
22.34
best: 27.96 (CASIA_IVA)
Image CaptioningonVizWiz 2020 test
CIDEr
56.19
best: 81.04 (IBM Research AI)
Image CaptioningonVizWiz 2020 test
METEOR
19.54
best: 22.25 (IBM Research AI)
Image CaptioningonVizWiz 2020 test
ROUGE-L
46.42
best: 50.2 (IBM Research AI)
Image CaptioningonVizWiz 2020 test
SPICE
14.72
best: 17.79 (CASIA_IVA)