List your company

Image Captioning on COCO (test)

154.9CIDErOFA
7.11645.48383.85122.217Jun 26, 2017Jun 21, 2018Jun 17, 2019Jun 12, 2020Jun 8, 2021Jun 4, 2022May 31, 2023

Evaluation results

MethodLinks
OFAModel Type=Autoregressive Model2023.05154.944.932.5———
CoDiModel Type=Diffusion Model2023.05149.940.231———
SotA2022.04149.6—————
BLIP2Model Type=Autoregressive Model2023.05145.843.7————
Pretrained FT SOTAFT=true, Shot=444K2022.04143.3—————
Flamingomode=Fine-tuned2022.04138.1—————
SCD-NetModel Type=Diffusion Model2023.05131.639.429.2———
OscarModel Type=Autoregressive Model2023.05124.1236.5830.4———
DDCapModel Type=Diffusion Model2023.05117.83528.2———
FlamingoFT=false, Shot=322022.04113.8—————
FlamingoFew-shot=32 shots2022.04113.8—————
ClipCapModel Type=Autoregressive Model2023.05108.3532.1527.1———
Flamingo-9BFT=false, Shot=322022.04106.3—————
FlamingoFT=false, Shot=42022.04103.2—————
Flamingo-3BFT=false, Shot=322022.0499—————
Flamingo-9BFT=false, Shot=42022.0493.1—————
Saliency and Context Attention2017.0689.828.424.852.1——
Flamingo-3BFT=false, Shot=42022.0485—————
FlamingoFT=false, Shot=02022.0484.3—————
Tavakoli et al.Saliency Map Model=GBVS2017.0684.128.723.551.2——
Tavakoli et al.Saliency Map Model=iSEEL2017.0683.628.323.550.8——
Flamingo-9BFT=false, Shot=02022.0479.4—————
LMCAPmode=few-shot2023.0575.919.922———
Flamingo-3BFT=false, Shot=02022.0473—————
Zero/Few shot SOTAFT=false, Shot=42022.0432.2—————
ZeroCapmode=zero-shot, language-specificity=English-specific2023.0514.62.611.5———
ConZICmode=zero-shot, language-specificity=English-specific2023.0512.81.311.5———
AnyDoorBackbone=BLIP-22025.11—40.731.660.9139.599.7
BadNetBackbone=BLIP-22025.11—40.531.760.9138.894.7
BlendedBackbone=BLIP-22025.11—40.931.661141.1100
CleanBackbone=BLIP-22025.11—42.531.961.8144.5—
CTPBackbone=BLIP-22025.11—40.831.560.9139.996.2
ShadowCastBackbone=BLIP-22025.11—40.631.760.9139.283.3
VLOODBackbone=BLIP-22025.11—39.930.860135.899.4