RaDL adds two attention modules to a Stable Diffusion based pipeline: Attribute Enhancement for per-instance attribute fidelity and Relation Attention that uses action verbs from the prompt to model inter-instance relationships.
Zero-shot text-to-image generation,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
RaDL: Relation-aware Disentangled Learning for Multi-Instance Text-to-Image Generation
RaDL adds two attention modules to a Stable Diffusion based pipeline: Attribute Enhancement for per-instance attribute fidelity and Relation Attention that uses action verbs from the prompt to model inter-instance relationships.