paper-with-me

Papers

Optimal Stepsize for Diffusion Sampling

2025-03-27 · Jianning Pei, Han Hu, Shuyang Gu

Diffusion models achieve remarkable generation quality but suffer from computational intensive sampling due to suboptimal step discretization. While existing works focus on optimizing denoising directions, we address the principled design of stepsize schedules. This paper proposes Optimal Stepsize Distillation, a dynamic programming framework that extracts theoretically optimal schedules by distilling knowledge from reference trajectories. By reformulating stepsize optimization as recursive error minimization, our method guarantees global discretization bounds through optimal substructure exploitation. Crucially, the distilled schedules demonstrate strong robustness across architectures, ODE solvers, and noise schedules. Experiments show 10x accelerated text-to-image generation while preserving 99.4% performance on GenEval. Our code is available at https://github.com/bebebe666/OptimalSteps.

📄 PDF Abstract BibTeX arXiv:2503.21774

Code (1)

bebebe666/optimalsteps 공식 구현 pytorch

Tasks

DenoisingImage GenerationText to Image GenerationText-to-Image Generation

Methods 이 논문이 사용한 방법론

Focus 설명 없음

Similar Papers 제목 키워드 기반

SGD: General Analysis and Improved Rates

2019-01-27 · Robert Mansel Gower, Nicolas Loizou, Xun Qian, Alibek Sailanbayev 외

We propose a general yet simple theorem describing the convergence of SGD under the arbitrary sampling paradigm. Our theorem describes the convergence of an infinite array of variants of SGD, each of which is associated …

Non-Asymptotic Guarantees for Average-Reward Q-Learning with Adaptive Stepsizes

2025-04-25 · Zaiwei Chen

This work presents the first finite-time analysis for the last-iterate convergence of average-reward Q-learning with an asynchronous implementation. A key feature of the algorithm we study is the use of adaptive stepsize…

Q-Learning

Bellman Optimal Stepsize Straightening of Flow-Matching Models

2023-12-27 · Bao Nguyen, Binh Nguyen, Viet Anh Nguyen

Flow matching is a powerful framework for generating high-quality samples in various applications, especially image synthesis. However, the intensive computational demands of these models, especially during the finetunin…

Image Generation

BELM: Bidirectional Explicit Linear Multi-step Sampler for Exact Inversion in Diffusion Models

2024-10-09 · Fangyikang Wang, Hubery Yin, Yuejiang Dong, Huminhao Zhu 외

The inversion of diffusion model sampling, which aims to find the corresponding initial noise of a sample, plays a critical role in various tasks. Recently, several heuristic exact inversion samplers have been proposed t…

From Score Matching to Diffusion: A Fine-Grained Error Analysis in the Gaussian Setting

2025-03-14 · Samuel Hurault, Matthieu Terris, Thomas Moreau, Gabriel Peyré

Sampling from an unknown distribution, accessible only through discrete samples, is a fundamental problem at the core of generative AI. The current state-of-the-art methods follow a two-step process: first, estimating th…

Denoising