SaaS, a self-adaptive attention-scaling method, improves instruction-following fidelity of unified image generation models without training by boosting the cross-attention activation of each sub-instruction in regions where the input image currently dominates.
Tam- ing Transformers for High-Resolution Image Synthesis
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
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Scale Your Instructions: Enhance the Instruction-Following Fidelity of Unified Image Generation Model by Self-Adaptive Attention Scaling
SaaS, a self-adaptive attention-scaling method, improves instruction-following fidelity of unified image generation models without training by boosting the cross-attention activation of each sub-instruction in regions where the input image currently dominates.