Input-Adaptive Generative Dynamics in Diffusion Models

ArXi:2411.15199v2 Announce Type: replace-cross Diffusion models typically generate data through a fixed denoising trajectory that is shared across all samples. However, generation targets can differ in complexity, suggesting that a single pre-defined diffusion process may not be optimal for every input. In this work, we investigate input-adaptive generative dynamics for diffusion models, where the generation process itself adapts to the conditions of each sample.