Mujie Lin

Mujie Lin

M.Phil. in Computer Science @ PKU. Building AI systems for molecular discovery, protein dynamics, and scientific foundation models.

Tikhonov-Stabilized Bézier Representation Forecasting for Training-free Diffusion Acceleration

Published in Conference on Neural Information Processing Systems (NeurIPS 2026), poster, 2026

  • Paper: TBD — add OpenReview / proceedings URL
  • Code: TBD — add GitHub repository URL

Abstract

Diffusion models, particularly Diffusion Transformers, achieve strong image and video generation quality but remain expensive at inference time due to repeated denoiser evaluations. Feature caching offers a deployment-friendly acceleration strategy by reusing or predicting intermediate representations without retraining the generator or modifying the sampler. However, existing cache-then-forecast methods often rely on Taylor-style finite-difference extrapolation, which can become unstable over longer cache intervals, and typically predict local module outputs whose errors may accumulate through subsequent denoiser blocks. We propose BeziCast, a training-free diffusion acceleration framework that forecasts output-proximal denoising representations using low-order Bézier trajectories. BeziCast estimates Bézier control points via Tikhonov-stabilized fitting, providing a smooth, capacity-controlled temporal parameterization that avoids explicit high-order derivative estimation and decouples trajectory capacity from cache-query density. To moderate aggressive extrapolation, we further derive equivalent forecasting weights and introduce a convex-hull-inspired guardrail that detects high-risk predictions and softly pulls them toward a conservative simplex-projected estimate. Extensive evaluations across advanced image and video diffusion models demonstrate the effectiveness of BeziCast. In particular, BeziCast accelerates FLUX.1 by up to $4.79\times$ and HunyuanVideo by up to $4.11\times$, while preserving substantially better generation quality than competing acceleration baselines.

Contribution

Co-first author (equal contribution). Contributed to idea design, implementation, experiment execution, theoretical proof, and paper writing.

#These authors contributed equally to this work. †Corresponding authors.

Recommended citation: Lei Zhu#, Mujie Lin#, Ruochong Zheng, Guangyi Wang, Hao Li, Peng Jin, Chang Liu†, Jie Chen†. "Tikhonov-Stabilized Bézier Representation Forecasting for Training-free Diffusion Acceleration." NeurIPS, 2026.
Download Paper