GRIT (zero-shot, no CBS, no VL pretraining, single model)

Reported on 2 benchmarks across 1 task · 1 paper

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

Natural Language Processing2 results

Image Captioningonnocaps out-of-domain
CIDEr· 2022-07-20
72.6
best: 126.67 (PaLI)
GRIT: Faster and Better Image captioning Transformer Using Dual Visual Features arXiv:2207.09666
Image Captioningonnocaps out-of-domain
SPICE· 2022-07-20
11.1
best: 15.7 (GIT, Single Model)
GRIT: Faster and Better Image captioning Transformer Using Dual Visual Features arXiv:2207.09666