woss

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-dev
B1
70.03
best: 72.48 (IBM Research AI)
Image CaptioningonVizWiz 2020 test-dev
B2
51.48
best: 53.93 (IBM Research AI)
Image CaptioningonVizWiz 2020 test-dev
B3
36.61
best: 39.28 (CASIA_IVA)
Image CaptioningonVizWiz 2020 test-dev
B4
25.8
best: 28.27 (CASIA_IVA)
Image CaptioningonVizWiz 2020 test-dev
CIDEr
70.78
best: 80.67 (IBM Research AI)
Image CaptioningonVizWiz 2020 test-dev
METEOR
21.06
best: 22.21 (IBM Research AI)
Image CaptioningonVizWiz 2020 test-dev
ROUGE-L
48.47
best: 50.27 (CASIA_IVA)
Image CaptioningonVizWiz 2020 test-dev
SPICE
16.4
best: 17.9 (CASIA_IVA)