DS-Flow: Dual-Stream Dynamical System Correction for Few-Step Diffusion Sampling

Diffusion models have shown remarkable success in generative modeling, but maintaining high quality at extremely low numbers of function evaluations (NFE) remains challenging due to large truncation errors under coarse ODE discretization. Existing methods either rely on costly distillation or perform step-level correction that adjusts the sampling direction only once per interval, which limits fine-grained trajectory refinement. In this work, we propose DS-Flow, which embeds an electronic dynamical-system (DS) corrector within each sampling step, enabling continuous in-step trajectory refinement through multiple hardware-efficient micro-steps without additional denoiser calls. While a PCA subspace is the natural choice for keeping each DS update lightweight, our analysis reveals that online PCA captures only $\sim$70% of truncation error; the residual exhibits spatial structure that a low-dimensional subspace cannot adequately represent. This motivates a dual-stream design: a global DS with dense coupling corrects the dominant low-dimensional error, while a local DS with grid coupling captures spatial residuals directly in pixel space. Extensive experiments demonstrate that DS-Flow consistently outperforms existing methods. For example, on CIFAR-10, DS-Flow reduces the Euler baseline's FID from 49.66 to 7.95 at NFE$=$5 while the corrector adds only $\sim$0.1% parameters. On DS hardware, the corrector runs at $\sim$mW power and $\sim \mu$s latency, achieving $\sim 10^4\times$ energy reduction and 6–7$\times$ speedup per correction interval over GPU-based alternatives.