Inference-time regularization to achieve Pareto-Optimal reward-diversity tradeoffs in text-to-image diffusion models.
Sep 6, 2024
