AI RESEARCH
Drifting Preference Optimization for One-Step Generative Models
arXiv CS.LG
•
ArXi:2606.02521v1 Announce Type: new One-step text-to-image generators are attractive for deployment because they generate an image with a single forward pass, but preference finetuning them remains difficult: standard alignment methods often rely on policy likelihoods, denoising trajectories, differentiable reward gradients, or test-time optimization. We propose Drifting Preference Optimization (DrPO), an online preference-finetuning method for deterministic one-step generators.