Si-Pei Zhao

DriftAudio: Marginal Drifting for Distributional Post-Training of One-Step Text-to-Audio Generators

Recent one-step text-to-audio (TTA) models substantially reduce inference cost, yet their generated distributions can still be improved through post-training. We propose DriftAudio, a distributional post-training method that adapts Drifting to pretrained one-step TTA generators. Applying Drifting condition-wise is chal...