English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Popular guidance for denoising diffusion probabilistic model (DDPM) linearlycombines distinct conditional models together to provide enhanced control oversamples. However, this approach overlooks nonlinear effects that becomesignificant when guidance scale is large. To address this issue, we proposecharacteristic guidance, a sampling method that provides first-principlenon-linear correction for classifier-free guided DDPMs. Such correction forcesthe guided DDPMs to respect the Fokker-Planck equation of their underlyingdiffusion process, in a way that is training-free, derivative-free, andcompatible with existing sampling methods. Experiments show that characteristicguidance enhances control and reduces color and exposure issues in imagegeneration, proving effective in diverse applications ranging from latent spacesampling to solving physics problems like magnet phase transitions.

Characteristic Guidance: Non-linear Correction for Diffusion Model at Large Guidance Scale