Cover Image

Fewer Denoising Steps or Cheaper Per-Step Inference: Towards Compute-Optimal Diffusion Model Deployment

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Du, Zhenbang, Fu, Yonggan, Wang, Lifu, Qian, Jiayi, Luo, Xiao, Yingyan, Lin
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps
by: Ma, Nanye, et al.
Published: (2025)

Denoising Diffusion Step-aware Models
by: Yang, Shuai, et al.
Published: (2023)

Early-Bird Diffusion: Investigating and Leveraging Timestep-Aware Early-Bird Tickets in Diffusion Models for Efficient Training
by: Whalen, Lexington, et al.
Published: (2025)

Step Saver: Predicting Minimum Denoising Steps for Diffusion Model Image Generation
by: Yu, Jean, et al.
Published: (2024)

Early-Bird GCNs: Graph-Network Co-Optimization Towards More Efficient GCN Training and Inference via Drawing Early-Bird Lottery Tickets
by: You, Haoran, et al.
Published: (2021)

DANCE: DAta-Network Co-optimization for Efficient Segmentation Model Training and Inference
by: Li, Chaojian, et al.
Published: (2021)

AccuQuant: Simulating Multiple Denoising Steps for Quantizing Diffusion Models
by: Lee, Seunghoon, et al.
Published: (2025)

DepthShrinker: A New Compression Paradigm Towards Boosting Real-Hardware Efficiency of Compact Neural Networks
by: Fu, Yonggan, et al.
Published: (2022)

Patch-Fool: Are Vision Transformers Always Robust Against Adversarial Perturbations?
by: Fu, Yonggan, et al.
Published: (2022)

Concept Unlearning by Modeling Key Steps of Diffusion Process
by: Zhang, Chaoshuo, et al.
Published: (2025)

FastLightGen: Fast and Light Video Generation with Fewer Steps and Parameters
by: Shao, Shitong, et al.
Published: (2026)

Gen-NeRF: Efficient and Generalizable Neural Radiance Fields via Algorithm-Hardware Co-Design
by: Fu, Yonggan, et al.
Published: (2023)

Omni-Recon: Harnessing Image-based Rendering for General-Purpose Neural Radiance Fields
by: Fu, Yonggan, et al.
Published: (2024)

NeRFool: Uncovering the Vulnerability of Generalizable Neural Radiance Fields against Adversarial Perturbations
by: Fu, Yonggan, et al.
Published: (2023)

Common Diffusion Noise Schedules and Sample Steps are Flawed
by: Lin, Shanchuan, et al.
Published: (2023)

Step-level Denoising-time Diffusion Alignment with Multiple Objectives
by: Zhang, Qi, et al.
Published: (2026)

Inference Compute-Optimal Video Vision Language Models
by: Wang, Peiqi, et al.
Published: (2025)

Text-to-Image Alignment in Denoising-Based Models through Step Selection
by: Grimal, Paul, et al.
Published: (2025)

Beyond Few-Step Inference: Accelerating Video Diffusion Transformer Model Serving with Inter-Request Caching Reuse
by: Liu, Hao, et al.
Published: (2026)

Learning Few-Step Diffusion Models by Trajectory Distribution Matching
by: Luo, Yihong, et al.
Published: (2025)

Single-Step Latent Diffusion for Underwater Image Restoration
by: Wu, Jiayi, et al.
Published: (2025)

SSDD-GAN: Single-Step Denoising Diffusion GAN for Cochlear Implant Surgical Scene Completion
by: Zhang, Yike, et al.
Published: (2025)

Gaussian Set Surface Reconstruction through Per-Gaussian Optimization
by: Huang, Zhentao, et al.
Published: (2025)

OSDFace: One-Step Diffusion Model for Face Restoration
by: Wang, Jingkai, et al.
Published: (2024)

AdaDiff: Accelerating Diffusion Models through Step-Wise Adaptive Computation
by: Tang, Shengkun, et al.
Published: (2023)

Auto-CARD: Efficient and Robust Codec Avatar Driving for Real-time Mobile Telepresence
by: Fu, Yonggan, et al.
Published: (2023)

Aligning Few-Step Diffusion Models with Dense Reward Difference Learning
by: Zhang, Ziyi, et al.
Published: (2024)

HonestFace: Towards Honest Face Restoration with One-Step Diffusion Model
by: Wang, Jingkai, et al.
Published: (2025)

Step-by-Step Diffusion: An Elementary Tutorial
by: Nakkiran, Preetum, et al.
Published: (2024)

Temporal Flexibility in Spiking Neural Networks: Towards Generalization Across Time Steps and Deployment Friendliness
by: Du, Kangrui, et al.
Published: (2025)

Diffusion Models Are Innate One-Step Generators
by: Zheng, Bowen, et al.
Published: (2024)

Another BRIXEL in the Wall: Towards Cheaper Dense Features
by: Lappe, Alexander, et al.
Published: (2025)

Efficient Diffusion Transformer with Step-wise Dynamic Attention Mediators
by: Pu, Yifan, et al.
Published: (2024)

Directly Denoising Diffusion Models
by: Zhang, Dan, et al.
Published: (2024)

AutoGAN-Distiller: Searching to Compress Generative Adversarial Networks
by: Fu, Yonggan, et al.
Published: (2020)

TDM-R1: Reinforcing Few-Step Diffusion Models with Non-Differentiable Reward
by: Luo, Yihong, et al.
Published: (2026)

Towards Multi-View Consistent Style Transfer with One-Step Diffusion via Vision Conditioning
by: Zuo, Yushen, et al.
Published: (2024)

Distilling ODE Solvers of Diffusion Models into Smaller Steps
by: Kim, Sanghwan, et al.
Published: (2023)

One-Step Diffusion Model for Image Motion-Deblurring
by: Liu, Xiaoyang, et al.
Published: (2025)

Clockwork Diffusion: Efficient Generation With Model-Step Distillation
by: Habibian, Amirhossein, et al.
Published: (2023)