List your company

Text-to-Image Retrieval on COCO-CN

78.7R@1M2-Encoder
12.24429.49746.7564.003Feb 14, 2022Oct 13, 2022Jun 12, 2023Feb 8, 2024Oct 7, 2024Jun 5, 2025Feb 2, 2026

Evaluation results

MethodLinks
M2-EncoderModel Size=10B, Protocol=Zero-shot2024.0178.79698.7———
M2-EncoderModel Size=1B, Protocol=Zero-shot2024.0175.995.198.3———
Wukong_ViT-LBackbone=ViT-L, Evaluation Protocol=Fine-tuned2022.027494.498.188.6——
M2-EncoderModel Size=0.4B, Protocol=Zero-shot2024.0172.593.797.7———
FILIP_ViT-LBackbone=ViT-L, Evaluation Protocol=Fine-tuned2022.0272.292.497.286.5——
Qwen3-VL-EmbeddingParameters=2B2026.0272.2—————
GMEParameters=7B2026.0271.8—————
CLIP_ViT-LBackbone=ViT-L, Evaluation Protocol=Fine-tuned2022.0270.192.296.486.2——
UME-R1Parameters=2B2026.0270.1—————
FILIP_Swin-LBackbone=Swin-L, Evaluation Protocol=Fine-tuned2022.0269.993.397.686.7——
ERNIE-ViL 2.0Setting=Zero-shot2022.0969.691.296.9—86.1—
ERNIE-ViL 2.0Protocol=Zero-shot2024.0169.691.296.9———
CN-CLIPModel Size=1B, Protocol=Zero-shot2024.0169.289.996.1———
CLIP_ViT-BBackbone=ViT-B, Evaluation Protocol=Fine-tuned2022.0268.993.397.386.6——
ObjEmbedParameters=4B2026.0268.8—————
FG-CLIP2Backbone=ViT-So/162026.0268.1—————
Wukong_ViT-BBackbone=ViT-B, Evaluation Protocol=Fine-tuned2022.026791.496.784.6——
CLIP_Swin-LBackbone=Swin-L, Evaluation Protocol=Fine-tuned2022.0266.791.596.385.4——
GMEParameters=2B2026.0266.7—————
ObjEmbedParameters=2B2026.0266.7—————
Wukong_Swin-LBackbone=Swin-L, Evaluation Protocol=Fine-tuned2022.026692.697.185.5——
MetaCLIP2Backbone=ViT-H/142026.0263.1—————
AltCLIPModel Size=0.8B, Protocol=Zero-shot2024.0161.38693.2———
VLM2Vec-V2Parameters=2B2026.0260.3—————
CN-CLIPModel Size=0.2B, Protocol=Zero-shot2024.015784.193.6———
R2D2Setting=Zero-shot2022.0956.48593.1—80.5—
R2D2Backbone=ViT-L, Zero-shot=true2022.0556.48593.1——80.5
R2D2Model Size=0.4B, Protocol=Zero-shot2024.0156.48593.1———
FILIP_ViT-BBackbone=ViT-B, Evaluation Protocol=Fine-tuned2022.0256.286.894.376.6——
TaisuModel Size=0.2B, Protocol=Zero-shot2024.0153.683.792.4———
WukongBackbone=ViT-L, Zero-shot=true2022.0253.480.290.175.1——
WukongSetting=Zero-shot2022.0953.480.290.1—75.1—
WukongBackbone=ViT-L, Zero-shot=true2022.0553.480.290.1——75.1
WukongModel Size=0.4B, Protocol=Zero-shot2024.0153.480.290.1———
FILIPBackbone=ViT-L, Zero-shot=true2022.0252.779.988.675.2——
FILIPBackbone=ViT-L, Zero-shot=true2022.0552.779.988.6——75.2
VLM2VecParameters=2B2026.0252.6—————
TaiyiSetting=Zero-shot2022.0951.58190.4———
SigLIP2Backbone=ViT-So/162026.0250.7—————
FILIPBackbone=Swin-L, Zero-shot=true2022.0250.579.288.672.1——
WukongBackbone=ViT-B, Zero-shot=true2022.0249.279.487.971.9——
CLIPBackbone=ViT-B, Zero-shot=true2022.0248.779.488.573——
CLIPBackbone=ViT-L, Zero-shot=true2022.0248.776.886.472.1——
CLIPBackbone=ViT-L, Zero-shot=true2022.0548.776.886.4——72.1
CLIPBackbone=Swin-L, Zero-shot=true2022.0246.778.187.771.7——
WukongBackbone=Swin-L, Zero-shot=true2022.0246.47787.670.8——
FILIPBackbone=ViT-B, Zero-shot=true2022.0237.568.18361.8——
BriVLBackbone=N/A, Zero-shot=true2022.0214.83954.237.4——
BriVLModel Size=1B, Protocol=Zero-shot2024.0114.83954.2———
M3PSetting=Zero-shot2022.09————32.3—