ETC: training-free diffusion models acceleration with Error-aware Trend Consistency
This work addresses the computational bottleneck in diffusion models for generative AI applications, offering an incremental improvement over existing training-free acceleration methods.
The paper tackles the slow iterative sampling of diffusion models by proposing ETC, a training-free acceleration framework that uses trend consistency and error tolerance to reduce trajectory deviations, achieving 2.65x speedup over FLUX with minimal quality loss.
Diffusion models have achieved remarkable generative quality but remain bottlenecked by costly iterative sampling. Recent training-free methods accelerate diffusion process by reusing model outputs. However, these methods ignore denoising trends and lack error control for model-specific tolerance, leading to trajectory deviations under multi-step reuse and exacerbating inconsistencies in the generated results. To address these issues, we introduce Error-aware Trend Consistency (ETC), a framework that (1) introduces a consistent trend predictor that leverages the smooth continuity of diffusion trajectories, projecting historical denoising patterns into stable future directions and progressively distributing them across multiple approximation steps to achieve acceleration without deviating; (2) proposes a model-specific error tolerance search mechanism that derives corrective thresholds by identifying transition points from volatile semantic planning to stable quality refinement. Experiments show that ETC achieves a 2.65x acceleration over FLUX with negligible (-0.074 SSIM score) degradation of consistency.