Key Features

Composes multiple image-generation capabilities through on-policy field distillation.
Treats each source capability as a frozen velocity field.
Uses hard-routed sample-wise field selection instead of soft averaging teachers.
Queries teacher fields on student-visited rollout states.
Uses one low-noise semantic query to reduce correlated supervision.
Trains with simple local velocity MSE matching.
Can absorb operator-defined fields such as classifier-free guidance.
Provides arXiv manuscript and public code link.

The method addresses target-field ambiguity, state-distribution mismatch, and trajectory-query correlation. It uses hard routing to select one frozen capability field per sample, a single low-noise semantic-side query on a stop-gradient student rollout state, and local velocity MSE matching.


DanceOPD is useful for unifying text-to-image generation, local editing, global editing, style absorption, and classifier-free-guidance absorption without degrading anchor capabilities. The project provides a manuscript, code link, diagnostics, ablations, and qualitative galleries.

Get more likes & reach the top of search results by adding this button on your site!

Embed button preview - Light theme
Embed button preview - Dark theme
TurboType Banner
Zero to AI Engineer Program

Zero to AI Engineer

Skip the degree. Learn real-world AI skills used by AI researchers and engineers. Get certified in 8 weeks or less. No experience required.

Subscribe to the AI Search Newsletter

Get top updates in AI to your inbox every weekend. It's free!