RNE: plug-and-play diffusion inference-time control and energy-based training

Understand

Diffusion models generate data by removing noise gradually, which corresponds to the time-reversal of a noising process.

• However, access to only the denoising kernels is often insufficient.
• In many applications, we need the knowledge of the marginal densities along the generation trajectory, which enables tasks such as inference-time control.
• To address this gap, in this paper, we introduce the Radon-Nikodym Estimator (RNE).