paper-with-me

홈 › Papers

Consistent Diffusion Meets Tweedie: Training Exact Ambient Diffusion Models with Noisy Data

2024-03-20 · Giannis Daras, Alexandros G. Dimakis, Constantinos Daskalakis

Ambient diffusion is a recently proposed framework for training diffusion models using corrupted data. Both Ambient Diffusion and alternative SURE-based approaches for learning diffusion models from corrupted data resort to approximations which deteriorate performance. We present the first framework for training diffusion models that provably sample from the uncorrupted distribution given only noisy training data, solving an open problem in this space. Our key technical contribution is a method that uses a double application of Tweedie's formula and a consistency loss function that allows us to extend sampling at noise levels below the observed data noise. We also provide further evidence that diffusion models memorize from their training sets by identifying extremely corrupted images that are almost perfectly reconstructed, raising copyright and privacy concerns. Our method for training using corrupted samples can be used to mitigate this problem. We demonstrate this by fine-tuning Stable Diffusion XL to generate samples from a distribution using only noisy samples. Our framework reduces the amount of memorization of the fine-tuning dataset, while maintaining competitive performance.

📄 PDF Abstract BibTeX arXiv:2404.10177

Code (1)

giannisdaras/ambient-tweedie 공식 구현 pytorch

Tasks

Memorization

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

Energy-Tweedie: Score meets Score, Energy meets Energy

2025-12-29 · Andrej Leban arxiv

Denoising and score estimation have long been known to be linked via the classical Tweedie's formula. In this work, we first extend the latter to a wider range of distributions often called "energy models" and denoted el…

Compositional Visual Planning via Inference-Time Diffusion Scaling

2026-03-03 · Yixin Zhang, Yunhao Luo, Utkarsh Aashu Mishra, Woo Chul Shin 외 arxiv

Diffusion models excel at short-horizon robot planning, yet scaling them to long-horizon tasks remains challenging due to computational constraints and limited training data. Existing compositional approaches stitch toge…

Tweedie's Formulae and Diffusion Generative Models Beyond Gaussian

2026-05-19 · Wenpin Tang, Nizar Touzi, Zikun Zhang, Xun Yu Zhou arxiv

Diffusion models have achieved remarkable success in generating samples from unknown data distributions. Most popular stochastic differential equation-based diffusion models perturb the target distribution by adding Gaus…

Gaussian Processes

SyncTweedies: A General Generative Framework Based on Synchronized Diffusions

2024-03-21 · Jaihoon Kim, Juil Koo, Kyeongmin Yeo, Minhyuk Sung

We introduce a general framework for generating diverse visual content, including ambiguous images, panorama images, mesh textures, and Gaussian splat textures, by synchronizing multiple diffusion processes. We present e…

Denoising

Binomial flows: Denoising and flow matching for discrete ordinal data

2026-05-01 · Yair Shenfeld, Ricardo Baptista, Stefano Peluchetti arxiv

Flow-based generative modeling in continuous spaces exploit Tweedie's formula to express the denoiser (learned in training) as a score function (used in sampling). In contrast, this relation has been largely missing in t…