paper-with-me

Papers

Morse: Dual-Sampling for Lossless Acceleration of Diffusion Models

2025-06-23 · Chao Li, Jiawei Fan, Anbang Yao

In this paper, we present Morse, a simple dual-sampling framework for accelerating diffusion models losslessly. The key insight of Morse is to reformulate the iterative generation (from noise to data) process via taking advantage of fast jump sampling and adaptive residual feedback strategies. Specifically, Morse involves two models called Dash and Dot that interact with each other. The Dash model is just the pre-trained diffusion model of any type, but operates in a jump sampling regime, creating sufficient space for sampling efficiency improvement. The Dot model is significantly faster than the Dash model, which is learnt to generate residual feedback conditioned on the observations at the current jump sampling point on the trajectory of the Dash model, lifting the noise estimate to easily match the next-step estimate of the Dash model without jump sampling. By chaining the outputs of the Dash and Dot models run in a time-interleaved fashion, Morse exhibits the merit of flexibly attaining desired image generation performance while improving overall runtime efficiency. With our proposed weight sharing strategy between the Dash and Dot models, Morse is efficient for training and inference. Our method shows a lossless speedup of 1.78X to 3.31X on average over a wide range of sampling step budgets relative to 9 baseline diffusion models on 6 image generation tasks. Furthermore, we show that our method can be also generalized to improve the Latent Consistency Model (LCM-SDXL, which is already accelerated with consistency distillation technique) tailored for few-step text-to-image synthesis. The code and models are available at https://github.com/deep-optimization/Morse.

📄 PDF Abstract BibTeX arXiv:2506.18251

Code (1)

deep-optimization/morse 공식 구현 pytorch

Tasks

Image Generation

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

FREE: Uncertainty-Aware Autoregression for Parallel Diffusion Transformers

2025-11-25 · Xinwan Wen, Bowen Li, Jiajun Luo, Ye Li 외 arxiv

Diffusion Transformers (DiTs) achieve state-of-the-art generation quality but require long sequential denoising trajectories, leading to high inference latency. Recent speculative inference methods enable lossless parall…

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

LinCa: Accelerating Diffusion Models via Learnable Decomposed Feature Caching

2026-08-18 · Jinshan Liu, Haoran Qin, Xiaobing Tu, Jiacheng Liu 외 arxiv

Diffusion models have achieved remarkable success in image and video generation, yet the high computational cost of iterative sampling remains a critical bottleneck for practical deployment. Feature caching has emerged a…

Video Generation

From Sketch to Fresco: Efficient Diffusion Transformer with Progressive Resolution

2026-01-12 · Shikang Zheng, Guantao Chen, Lixuan He, Jiacheng Liu 외 arxiv

Diffusion Transformers achieve impressive generative quality but remain computationally expensive due to iterative sampling. Recently, dynamic resolution sampling has emerged as a promising acceleration technique by redu…

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…