IV-CoT introduces an implicit chain-of-thought framework that decomposes visual queries into a structural-to-semantic cascade with training-only sketch supervision to improve structure-aware text-to-image generation.
arXiv preprint arXiv:2510.13500 , year=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
IV-CoT: Implicit Visual Chain-of-Thought for Structure-Aware Text-to-Image Generation
IV-CoT introduces an implicit chain-of-thought framework that decomposes visual queries into a structural-to-semantic cascade with training-only sketch supervision to improve structure-aware text-to-image generation.