paper-with-me

홈 › Papers

Forecast then Calibrate: Feature Caching as ODE for Efficient Diffusion Transformers

2025-08-22 · Shikang Zheng, Liang Feng, Xinyu Wang, Qinming Zhou, Peiliang Cai, Chang Zou, Jiacheng Liu, Yuqi Lin, Junjie Chen, Yue Ma, Linfeng Zhang arxiv

Diffusion Transformers (DiTs) have demonstrated exceptional performance in high-fidelity image and video generation. To reduce their substantial computational costs, feature caching techniques have been proposed to accelerate inference by reusing hidden representations from previous timesteps. However, current methods often struggle to maintain generation quality at high acceleration ratios, where prediction errors increase sharply due to the inherent instability of long-step forecasting. In this work, we adopt an ordinary differential equation (ODE) perspective on the hidden-feature sequence, modeling layer representations along the trajectory as a feature-ODE. We attribute the degradation of existing caching strategies to their inability to robustly integrate historical features under large skipping intervals. To address this, we propose FoCa (Forecast-then-Calibrate), which treats feature caching as a feature-ODE solving problem. Extensive experiments on image synthesis, video generation, and super-resolution tasks demonstrate the effectiveness of FoCa, especially under aggressive acceleration. Without additional training, FoCa achieves near-lossless speedups of 5.50 times on FLUX, 6.45 times on HunyuanVideo, 3.17 times on Inf-DiT, and maintains high quality with a 4.53 times speedup on DiT.

📄 PDF Abstract BibTeX arXiv:2508.16211

Code (0)

등록된 구현이 없습니다.

Tasks

Video Generation

Similar Papers 제목 키워드 기반

Let Features Decide Their Own Solvers: Hybrid Feature Caching for Diffusion Transformers

2025-10-05 · Shikang Zheng, Guantao Chen, Qinming Zhou, Yuqi Lin 외 arxiv

Diffusion Transformers offer state-of-the-art fidelity in image and video synthesis, but their iterative sampling process remains a major bottleneck due to the high cost of transformer forward passes at each timestep. To…

From Reusing to Forecasting: Accelerating Diffusion Models with TaylorSeers

2025-03-10 · Jiacheng Liu, Chang Zou, Yuanhuiyi Lyu, Junjie Chen 외

Diffusion Transformers (DiT) have revolutionized high-fidelity image and video synthesis, yet their computational demands remain prohibitive for real-time applications. To solve this problem, feature caching has been pro…

Accelerating Diffusion Transformer via Increment-Calibrated Caching with Channel-Aware Singular Value Decomposition

2025-05-09 · CVPR 2025 1 · ZhiYuan Chen, Keyi Li, Yifan Jia, Le Ye 외

Diffusion transformer (DiT) models have achieved remarkable success in image generation, thanks for their exceptional generative capabilities and scalability. Nonetheless, the iterative nature of diffusion models (DMs) r…

Image Generation

Forecast the Principal, Stabilize the Residual: Subspace-Aware Feature Caching for Efficient Diffusion Transformers

2026-01-12 · Guantao Chen, Shikang Zheng, Yuqi Lin, Linfeng Zhang arxiv

Diffusion Transformer (DiT) models have achieved unprecedented quality in image and video generation, yet their iterative sampling process remains computationally prohibitive. To accelerate inference, feature caching met…

Video Generation

Adaptive Spectral Feature Forecasting for Diffusion Sampling Acceleration

2026-03-02 · Jiaqi Han, Juntong Shi, Puheng Li, Haotian Ye 외 arxiv

Diffusion models have become the dominant tool for high-fidelity image and video generation, yet are critically bottlenecked by their inference speed due to the numerous iterative passes of Diffusion Transformers. To red…

Video Generation