Image Classification on MLRSNet
81.1AccuracySOTA
Evaluation Results
| Method | Links | |
|---|---|---|
| SOTASetting=Transductive, Transductive Inference (TransCLIP)=false2025.06 | 81.1 | |
| GeoMELTBackbone=BEiT-3, zero-shot=true2025.12 | 73.21 | |
| T-SkyCLIP50Setting=Transductive, Transductive Inference (TransCLIP)=true2025.06 | 73.2 | |
| T-GeoRSCLIPSetting=Transductive, Transductive Inference (TransCLIP)=true2025.06 | 71.9 | |
| MetaCLIP-2.5BArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 70.7 | |
| GeoRSCLIPARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 70.2 | |
| SigLIP2Architecture=ViT-L/16, Setting=Zero-shot, Template=DST2026.06 | 70.2 | |
| SkyCLIPARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 68.9 | |
| SkyCLIPArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 68.3 | |
| SigLIP2ARCH=ViT-L/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 68.1 | |
| GeoRSCLIPArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 67.8 | |
| MetaCLIP-2.5BARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 66.7 | |
| RemoteCLIPBackbone=ViT-L-14, zero-shot=true2023.06 | 66.32 | |
| RemoteCLIPBackbone=ViT-L-14, zero-shot=true2023.06 | 66.32 | |
| CLIP LAION-RSArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 66.1 | |
| T-RemoteCLIPSetting=Transductive, Transductive Inference (TransCLIP)=true2025.06 | 65.8 | |
| GeoRSCLIPARCH=ViT-B/32, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 65.7 | |
| CLIPArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 65.6 | |
| GeoRSCLIPSetting=Transductive, Transductive Inference (TransCLIP)=false2025.06 | 65 | |
| CLIPBackbone=ViT-L-14, zero-shot=true2023.06 | 64.89 | |
| CLIPBackbone=ViT-L-14, zero-shot=true2023.06 | 64.89 | |
| SigLIPARCH=ViT-L/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 64.7 | |
| MetaCLIP-2.5BArchitecture=ViT-B/16, Setting=Zero-shot, Template=DST2026.06 | 64.4 | |
| T-CLIPSetting=Transductive, Transductive Inference (TransCLIP)=true2025.06 | 64.2 | |
| GeoRSCLIPArchitecture=ViT-B/32, Setting=Zero-shot, Template=DST2026.06 | 64.1 | |
| CLIP LAION-RSARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 63.5 | |
| SkyCLIP50Setting=Transductive, Transductive Inference (TransCLIP)=false2025.06 | 63.2 | |
| CLIPARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 62.7 | |
| MetaCLIP-400MArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 62.7 | |
| MetaCLIP-400MARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 62.6 | |
| SigLIP2ARCH=ViT-B/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 62.5 | |
| SigLIP2Architecture=ViT-B/16, Setting=Zero-shot, Template=DST2026.06 | 62.5 | |
| MetaCLIP-2.5BARCH=ViT-B/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 62.3 | |
| RemoteCLIPBackbone=ViT-L, zero-shot=true2025.12 | 61.67 | |
| RemoteCLIPArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 61.6 | |
| SigLIPArchitecture=ViT-L/16, Setting=Zero-shot, Template=DST2026.06 | 60.8 | |
| RemoteCLIPARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 60.7 | |
| MetaCLIP-400MArchitecture=ViT-B/16, Setting=Zero-shot, Template=DST2026.06 | 59.8 | |
| MetaCLIP-400MARCH=ViT-B/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 59.3 | |
| RemoteCLIPBackbone=ViT-B-32, zero-shot=true2023.06 | 59.28 | |
| RemoteCLIPBackbone=ViT-B-32, zero-shot=true2023.06 | 59.28 | |
| CLIPBackbone=ViT-L, zero-shot=true2025.12 | 59.09 | |
| SigLIPARCH=ViT-B/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 58.3 | |
| RemoteCLIPARCH=ViT-B/32, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 58.2 | |
| RemoteCLIPBackbone=ViT-B, zero-shot=true2025.12 | 57.91 | |
| CLIPArchitecture=ViT-B/16, Setting=Zero-shot, Template=DST2026.06 | 57.2 | |
| RemoteCLIPArchitecture=ViT-B/32, Setting=Zero-shot, Template=DST2026.06 | 56.9 | |
| SigLIPArchitecture=ViT-B/16, Setting=Zero-shot, Template=DST2026.06 | 56.7 | |
| CLIPARCH=ViT-B/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 56.5 | |
| RemoteCLIPSetting=Transductive, Transductive Inference (TransCLIP)=false2025.06 | 56.3 | |
| CLIPBackbone=ViT-B-32, zero-shot=true2023.06 | 55.29 | |
| CLIPBackbone=ViT-B-32, zero-shot=true2023.06 | 55.29 | |
| CLIPSetting=Transductive, Transductive Inference (TransCLIP)=false2025.06 | 51.2 | |
| CLIPBackbone=ResNet-50, zero-shot=true2023.06 | 45.54 | |
| CLIPBackbone=ResNet-50, zero-shot=true2023.06 | 45.54 | |
| RemoteCLIPARCH=RN-50, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 45.2 | |
| RemoteCLIPArchitecture=RN-50, Setting=Zero-shot, Template=DST2026.06 | 42.3 | |
| CLIPBackbone=ViT-B, zero-shot=true2025.12 | 41.12 | |
| RemoteCLIPBackbone=ResNet-50, zero-shot=true2023.06 | 40.68 | |
| RemoteCLIPBackbone=ResNet-50, zero-shot=true2023.06 | 40.68 |