AI RESEARCH

Elucidating the SNR-t Bias of Diffusion Probabilistic Models

arXiv CS.CV

ArXi:2604.16044v1 Announce Type: new Diffusion Probabilistic Models have nstrated remarkable performance across a wide range of generative tasks. However, we have observed that these models often suffer from a Signal-to-Noise Ratio-timestep (SNR-t) bias. This bias refers to the misalignment between the SNR of the denoising sample and its corresponding timestep during the inference phase. Specifically, during