Text-to-Image Generation on Flickr30k
54.33FIDSD v3.5
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SD v3.5Mode=frozen2026.06 | 54.33 | 32.66 | |
| ELDiffBackbone=SD v3.5, Finetuning Strategy=ft. w. Ours2026.06 | 54.79 | 32.78 | |
| IterCompBackbone=SD v3.5, Finetuning Strategy=ft. w. IterComp2026.06 | 54.85 | 32.52 | |
| TokenComposeBackbone=SD v3.5, Finetuning Strategy=ft. w. TokenCompose2026.06 | 54.87 | 32.87 | |
| CoMatBackbone=SD v3.5, Finetuning Strategy=ft. w. CoMat2026.06 | 55.01 | 32.93 | |
| ELDiffBackbone=SD v2.1, Finetuning Strategy=ft. w. Ours2026.06 | 55.12 | 0.3348 | |
| LLDMBackbone=SD v3.5, Finetuning Strategy=ft. w. LLDM2026.06 | 55.14 | 32.43 | |
| IterCompBackbone=SD v2.1, Finetuning Strategy=ft. w. IterComp2026.06 | 56.31 | 0.3237 | |
| TokenComposeBackbone=SD v2.1, Finetuning Strategy=ft. w. TokenCompose2026.06 | 56.43 | 0.3264 | |
| CoMatBackbone=SD v2.1, Finetuning Strategy=ft. w. CoMat2026.06 | 56.52 | 0.3185 | |
| SD v2.1Mode=frozen2026.06 | 56.83 | 0.3159 | |
| LLDMBackbone=SD v2.1, Finetuning Strategy=ft. w. LLDM2026.06 | 57.02 | 0.3231 | |
| SD-2.1-baseResolution=512x5122025.09 | 202.6 | 32.71 | |
| SD-1.5Resolution=512x5122025.09 | 203.8 | 32.31 | |
| MaskAttn-SDXLSetting=mid-res, default, Resolution=1024x10242025.09 | 206.98 | 33.54 | |
| PixArt-ΣInference Setting=method-recommended2025.09 | 209.12 | 32.01 | |
| SDXLResolution=1024x10242025.09 | 209.8 | 33.03 | |
| PixArt-αInference Setting=method-recommended2025.09 | 210.3 | 32.12 |