A data-curation engine plus a token-order attention injection module raises spatial accuracy of Stable Diffusion and FLUX models on standard benchmarks.
Cohn, Dayou Liu, Sheng-Sheng Wang, Jihong Ouyang, and Qiangyuan Yu
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
CoMPaSS: Enhancing Spatial Understanding in Text-to-Image Diffusion Models
A data-curation engine plus a token-order attention injection module raises spatial accuracy of Stable Diffusion and FLUX models on standard benchmarks.