Text-to-image generation on TIFA
84.63TIFADPT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DPTStable Diffusion version=v2.1, Self-Correction (SC)=true2024.03 | 84.63 | — | |
| DPTStable Diffusion version=v2.1, Self-Correction (SC)=false2024.03 | 84.49 | — | |
| DPTStable Diffusion version=v1.4, Self-Correction (SC)=true2024.03 | 82.4 | — | |
| DPTStable Diffusion version=v1.4, Self-Correction (SC)=false2024.03 | 82.04 | — | |
| HN-DiffusionITMStable Diffusion version=v2.1, Self-Correction (SC)=false2024.03 | 82.02 | — | |
| Attend-and-ExciteStable Diffusion version=v2.1, Self-Correction (SC)=false2024.03 | 81.98 | — | |
| SD-v2.1Stable Diffusion version=v2.1, Self-Correction (SC)=false2024.03 | 81.35 | — | |
| UIO-2model_size=XXL2023.12 | 81.3 | — | |
| Stable Diffusion v2-1-baseNFE=50, Retrieval=false2026.02 | 81.1 | 6.04 | |
| ImageRAGTurboNFE=1, Retrieval=true2026.02 | 80.1 | 5.88 | |
| UIO-2model_size=XL2023.12 | 80 | — | |
| minDALL-E2023.12 | 79.4 | — | |
| LayoutGPTStable Diffusion version=v1.4, Self-Correction (SC)=false2024.03 | 79.31 | — | |
| SD-v1.4Stable Diffusion version=v1.4, Self-Correction (SC)=false2024.03 | 79.15 | — | |
| HN-DiffusionITMStable Diffusion version=v1.4, Self-Correction (SC)=false2024.03 | 79.02 | — | |
| UIO-2XXLzero-shot=true, size=XXL2023.12 | 78.7 | — | |
| SD v1.5zero-shot=true2023.12 | 78.4 | — | |
| SD-1.52023.12 | 78.4 | — | |
| Stable Diffusion Turbo v2-1-baseNFE=1, Retrieval=false2026.02 | 77.9 | 5.85 | |
| VPGenStable Diffusion version=v1.4, Self-Correction (SC)=false2024.03 | 77.33 | — | |
| UIO-2XLzero-shot=true, size=XL2023.12 | 77.2 | — | |
| Stable Diffusion v1-5NFE=50, Retrieval=false2026.02 | 76.8 | 5.79 | |
| Latent Consistency ModelNFE=4, Retrieval=false2026.02 | 76.4 | 5.8 | |
| UIO-2model_size=L2023.12 | 74.3 | — | |
| RDMNFE=50, Retrieval=true2026.02 | 72.5 | 5.4 | |
| CoDi2023.12 | 71.6 | — | |
| UIO-2Lzero-shot=true, size=L2023.12 | 70.2 | — | |
| Emu2023.12 | 65.5 | — |