paper-with-me

Papers

Nonequilbrium physics of generative diffusion models

2024-05-20 · Zhendong Yu, Haiping Huang

Generative diffusion models apply the concept of Langevin dynamics in physics to machine leaning, attracting a lot of interests from engineering, statistics and physics, but a complete picture about inherent mechanisms is still lacking. In this paper, we provide a transparent physics analysis of diffusion models, formulating the fluctuation theorem, entropy production, equilibrium measure, and Franz-Parisi potential to understand the dynamic process and intrinsic phase transitions. Our analysis is rooted in a path integral representation of both forward and backward dynamics, and in treating the reverse diffusion generative process as a statistical inference, where the time-dependent state variables serve as quenched disorder akin to that in spin glass theory. Our study thus links stochastic thermodynamics, statistical inference and geometry based analysis together to yield a coherent picture about how the generative diffusion models work.

📄 PDF Abstract BibTeX arXiv:2405.11932

Code (0)

등록된 구현이 없습니다.

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

Physics-Inspired Generative Models in Medical Imaging: A Review

2024-07-15 · Dennis Hein, Afshin Bozorgpour, Dorit Merhof, Ge Wang

Physics-inspired Generative Models (GMs), in particular Diffusion Models (DMs) and Poisson Flow Models (PFMs), enhance Bayesian methods and promise great utility in medical imaging. This review examines the transformativ…

DenoisingImage GenerationImage Reconstruction

Generating Synthetic Net Load Data with Physics-informed Diffusion Model

2024-06-04 · Shaorong Zhang, Yuanbin Cheng, Nanpeng Yu

This paper presents a novel physics-informed diffusion model for generating synthetic net load data, addressing the challenges of data scarcity and privacy concerns. The proposed framework embeds physical models within d…

DenoisingDiversity

Unified framework for diffusion generative models in SO(3): applications in computer vision and astrophysics

2023-12-18 · Yesukhei Jagvaral, Francois Lanusse, Rachel Mandelbaum

Diffusion-based generative models represent the current state-of-the-art for image generation. However, standard diffusion models are based on Euclidean geometry and do not translate directly to manifold-valued data. In …

AstronomyDenoisingImage GenerationPose Estimation

Particle-Guided Diffusion Models for Partial Differential Equations

2026-01-30 · Andrew Millard, Fredrik Lindsten, Zheng Zhao arxiv

We introduce a guided stochastic sampling method that augments sampling from diffusion models with physics-based guidance derived from partial differential equation (PDE) residuals and observational constraints, ensuring…

PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation

2025-09-24 · Chen Wang, Chuhao Chen, Yiming Huang, Zhiyang Dou 외 arxiv

Existing video generation models excel at producing photo-realistic videos from text or images, but often lack physical plausibility and 3D controllability. To overcome these limitations, we introduce PhysCtrl, a novel f…

Video Generation