IV-CoT: Implicit Visual Chain-of-Thought for Structure-Aware Text-to-Image Generation
Open paper
Benchmark
Model
Rank
Results
text-to-image-generation-on-geneval
Full IV-CoT (OpenUni-L-1024, 512 queries)
–
Overall: 0.885
text-to-image-generation-on-t2i-compbench
IV-CoT (OpenUni-L-1024)
–
Color: 0.8542
Texture: 0.755
Spatial: 0.4734
Shape: 0.6296
…
Rank counts only results with a code link.