MixtureTT performs direct per-stem timbre transfer on polyphonic mixtures via a shared diffusion transformer, outperforming single-stem baselines on SATB choral data while eliminating cascaded separation errors.
High-resolution image synthesis with latent diffusion models
2 Pith papers cite this work. Polarity classification is still indexing.
citation-role summary
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
A windowed cross-attention control method on skip features enables geometry-controlled high-resolution satellite image synthesis from pre-trained diffusion models with better alignment to control maps than prior techniques.
citing papers explorer
-
Remix the Timbre: Diffusion-Based Style Transfer Across Polyphonic Stems
MixtureTT performs direct per-stem timbre transfer on polyphonic mixtures via a shared diffusion transformer, outperforming single-stem baselines on SATB choral data while eliminating cascaded separation errors.
-
Efficient Geometry-Controlled High-Resolution Satellite Image Synthesis
A windowed cross-attention control method on skip features enables geometry-controlled high-resolution satellite image synthesis from pre-trained diffusion models with better alignment to control maps than prior techniques.