AI RESEARCH
Accelerating Frequency Domain Diffusion Models with Error-Feedback Event-Driven Caching
arXiv CS.LG
•
ArXi:2604.22901v1 Announce Type: new Diffusion models achieve remarkable success in time series generation. However, slow inference limits their practical deployment. We propose E$^2$-CRF (Error-Feedback Event-Driven Cumulative Residual Feature caching) to accelerate frequency domain diffusion models. Our method exploits two structural properties: (1) spectral localization, where signal energy concentrates in low frequencies, and (2) mirror symmetry, which halves the effective frequency dimension.