List your company

Text-to-image Retrieval on Flickr30k (1K)

84.9R@1ALIGN
-3.29219.60442.565.396Feb 5, 2021Dec 14, 2021Oct 23, 2022Sep 1, 2023Jul 10, 2024May 19, 2025Mar 28, 2026

Evaluation results

MethodLinks
ALIGN#Images=1.2B, Fine-tuning=true2022.0284.997.498.6
Triple Contrastive Learning (TCL)#Images=4M, Fine-tuning=true2022.028496.798.5
ALBEF#Images=4M, Fine-tuning=true2022.0282.896.798.4
SoftMask++#Img=4M, Zero-shot=true2023.0480.194.997.7
CODIS#Img=4M, Zero-shot=true2023.0479.794.897.3
TCL#Images=4M, Zero-shot=true2022.0279.695.197.4
TCL#Img=4M, Zero-shot=true2023.0479.695.197.4
ALBEF#Images=4M, Zero-shot=true2022.0276.893.796.7
ALBEF#Img=4M, Zero-shot=true2023.0476.893.796.7
VILLA#Images=4M, Fine-tuning=true2022.0276.394.296.8
VISTA-LTime (ms)=~40, Fine-tuning=true2022.0375.894.296.9
ALIGN#Images=1.2B, Zero-shot=true2022.0275.793.896.8
ALIGN#Img=1.2B, Zero-shot=true2023.0475.793.896.8
UNITER#Images=4M, Fine-tuning=true2022.0275.694.196.8
UNIMO#Images=4M, Fine-tuning=true2022.0274.793.4796.1
ERNIE-VIL-BTime (ms)=~920, Fine-tuning=true2022.0374.492.795.9
VSEinftyFine-tuning=true2022.0374.293.796.8
H Xue et al.Fine-tuning=true2022.0373.593.196.4
ImageBERT#Images=6M, Fine-tuning=true2022.0273.192.696
SOHO#Images=200K, Fine-tuning=true2022.0272.592.796.1
UNITER-BTime (ms)=~900, Fine-tuning=true2022.0372.592.496.1
SOHOFine-tuning=true2022.0372.592.796.1
Miech et alFine-tuning=true2022.0372.191.595.2
Unicoder-VLTime (ms)=925, Fine-tuning=true2022.0371.591.295.2
Pixel-BERT-XTime (ms)=~160, Fine-tuning=true2022.0371.592.195.8
VISTA-BTime (ms)=~17, Fine-tuning=true2022.0368.991.195.1
CLIP#Images=400M, Zero-shot=true2022.0268.790.695.2
CLIP#Img=400M, Zero-shot=true2023.0468.790.695.2
12-in-1Fine-tuning=true2022.0367.989.694.2
UNITER-BaseVisual Embed=Region, Time (ms)=~900, Zero-Shot=true2021.0266.288.492.9
UNITER#Images=4M, Zero-shot=true2022.0266.288.492.9
UNITER#Img=4M, Zero-shot=true2023.0466.288.492.9
ViLT#Images=4M, Fine-tuning=true2022.0264.488.793.8
ViLT-BTime (ms)=~15, Fine-tuning=true2022.0364.488.793.8
CLIP ViT-B/32|S|=400M2026.0358.883.490
SGRAFFine-tuning=true2022.0358.58388.8
GSMNFine-tuning=true2022.0357.482.389
ViLT-B/32Visual Embed=Linear, Time (ms)=~15, Zero-Shot=true, Pre-training steps=200K2021.025582.589.8
ViLT#Images=4M, Zero-shot=true2022.025582.589.8
ViLT#Img=4M, Zero-shot=true2023.045582.589.8
VSRNFine-tuning=true2022.0354.781.888.2
Procrustes|S|=5002026.0354.469.578
ImageBERTVisual Embed=Region, Time (ms)=~925, Zero-Shot=true, Additional Pre-training Dataset=10M proprietary2021.0254.379.687.5
ImageBERT#Images=6M, Zero-shot=true2022.0254.379.687.5
IMRAMFine-tuning=true2022.0353.979.487.2
Pixel-BERT-RTime (ms)=~60, Fine-tuning=true2022.0353.480.488.5
ViLT-B/32Visual Embed=Linear, Time (ms)=~15, Zero-Shot=true2021.0251.379.987.9
SCANFine-tuning=true2022.0348.677.785.2
Unicoder-VLVisual Embed=Region, Time (ms)=~925, Zero-Shot=true2021.0248.47685.2
ViLBERTVisual Embed=Region, Time (ms)=~900, Zero-Shot=true2021.0231.961.172.8
Relative Reps|S|=5002026.0326.750.962.3
Relative Reps|S|=1002026.0312.127.536.7
Procrustes|S|=1002026.0310.82332.9
FMap|S|=5002026.036.117.925.9
Relative Reps|S|=202026.033.59.314.1
Procrustes|S|=202026.0325.710.4
FMap|S|=1002026.031.99.315
FMap|S|=202026.030.84.58.1
FMap Unsupervised (HKS)|S|=02026.030.41.73.2
Raw Cosine|S|=02026.030.10.51
CCA|S|=202026.030.10.41
CCA|S|=1002026.030.10.50.9
CCA|S|=5002026.030.10.40.9