Latent Fourier Transform

ICLROral2026

Authors: Mason Long Wang, Cheng-Zhi Anna Huang
Affiliation: Massachusetts Institute of Technology
Venue: ICLR 2026
Track: Oral

TL;DR

We introduce novel frequency-domain controls for generative music models by applying the Fourier transform to the latent space of a diffusion autoencoder.

Opening excerpt from the authors’ abstract. source

Read the paper

OpenReview

Topics

diffusion control

← All ICLR 2026 Oral papers · Browse the whole archive