List your company

Scene Text Visual Question Answering on ST-VQA (val)

0.845ANLSPaLI-X
0.181480.353740.5260.69826May 27, 2022Jul 27, 2022Sep 26, 2022Nov 26, 2022Jan 26, 2023Mar 28, 2023May 29, 2023

Evaluation results

MethodLinks
PaLI-XOCR pipeline input=true2023.050.845—
PaLIOCR pipeline input=true2023.050.799—
PaLI-XOCR pipeline input=false2023.050.799—
PaLI-17BVocabulary setting=Open-vocabulary generation2022.090.771—
GIT2OCR pipeline input=false2023.050.758—
GIT2Parameters=5.1B, Vocabulary setting=Closed-vocabulary classification2022.090.751—
GIT2Parameters=5.1B2022.050.75166.6
PaLI-15BVocabulary setting=Open-vocabulary generation2022.090.732—
LaTr2022.050.70261.64
GITParameters=0.7B, Vocabulary setting=Closed-vocabulary classification2022.090.691—
GITParameters=0.7B2022.050.69159.2
PaLI-3BVocabulary setting=Open-vocabulary generation2022.090.675—
TAP + TAGExtra Data=TextVQA2022.080.620.5353
TAP††Extra Data=TextVQA, TextCaps, OCR-CC2022.080.5980.5083
TAP2022.050.59850.8
TAP + TAGExtra Data=None2022.080.5950.5018
LOGOSExtra Data=TextVQA2022.080.5810.4863
M4C⁺ + TAGExtra Data=TextVQA2022.080.5790.4869
M4C†Extra Data=TextVQA2022.080.560.466
TAPExtra Data=None2022.080.5510.4529
M4C⁺ + TAGExtra Data=None2022.080.540.4452
LOGOSExtra Data=None2022.080.5350.441
M4C†Extra Data=None2022.080.5170.4228
SA-M4CExtra Data=None2022.080.5120.4223
SA-M4C2022.050.51242.2
LaAP-NetExtra Data=None2022.080.4970.3974
LaAP-Net2022.050.49739.7
M4C2022.050.47238.1
GITLParameters=0.3B2022.050.44632.3
GITBParameters=0.1B2022.050.20714.7