Flow Matching (FM) (also referred to as stochastic interpolants or rectified flows) stands out as a class of generative models that aims to bridge in finite time the target distribution $ν^\star$ with an auxiliary distribution $μ$, leveraging a fixed coupling $π$ and a bridge which can either be deterministic or stochastic. These two ingredients define a path measure which can then be approximated by learning the drift of its Markovian projection. The main contribution of this paper is to provide relatively mild assumptions on $ν^\star$, $μ$ and $π$ to obtain non-asymptotics guarantees for Diffusion Flow Matching (DFM) models using as bridge the conditional distribution associated with the Brownian motion. More precisely, we establish bounds on the Kullback-Leibler divergence between the target distribution and the one generated by such DFM models under moment conditions on the score of $ν^\star$, $μ$ and $π$, and a standard $L^2$-drift-approximation error assumption.
Theoretical Guarantees for SMC-Guided Diffusion Sampling
Sorbonne Université and Université Paris Cité, CNRS, LPSM(索邦大学和巴黎大学,法国国家科学研究中心,巴黎高等师范学院);Center for Statistics and Images, Mines Paris, PSL University(统计与图像中心,巴黎 Mines,PSL 大学);CREST, ENSAE Paris, Institut Polytechnique de Paris(CREST,ENSAE 巴黎,巴黎理工学院)
Alignment of Diffusion Model and Flow Matching for Text-to-Image Generation
Department of Statistics, University of California, Los Angeles(加州大学洛杉矶分校统计学系);Department of Industrial and Systems Engineering, University of Minneasota(明尼苏达大学工业与系统工程系);School of Data Science, The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)数据科学学院)
Improving Probabilistic Diffusion Models With Optimal Diagonal Covariance Matching
Imperial College London(伦敦帝国学院);University College London(伦敦大学学院);University of Manchester(曼彻斯特大学);Max Planck Institute for Intelligent Systems(马克斯·普朗克智能系统研究所);University of Tubingen(蒂宾根大学);IMPRS-IS(马克斯·普朗克智能系统研究所国际马克斯·普朗克研究学院)
Center for Data Science(数据科学中心);Peking University(北京大学);Guanghua School of Management(光华管理学院);Department of Statistics and Data Science(统计与数据科学系);Tsinghua University(清华大学)
Soft Diffusion: Score Matching for General Corruptions
Google Research(谷歌研究院);UT Austin(得克萨斯大学奥斯汀分校)
我们定义了一个更广泛的家庭腐败过程,概括了以前已知的扩散模型。为了扭转这些一般的扩散,我们提出了一个新的目标,称为软分数匹配,可证明学习任何线性腐败过程的分数函数,并为CelebA产生最先进的结果。软分数匹配结合了网络中的退化过程。我们的新损失训练模型预测一个干净的图像,\textit{that after corruption},匹配扩散观察。我们表明,我们的目标学习梯度的可能性在适当的规律性条件下的一个家庭的腐败过程。我们进一步开发了一种原则性的方法来选择一般扩散过程的腐败水平和一种新的采样方法,我们称之为动量采样器。我们的实验表明,我们的框架适用于一般的线性腐败过程,如高斯模糊和掩蔽。我们在CelebA-64上实现了最先进的FID评分$1.85$,优于所有先前的线性扩散模型。与香草去噪扩散相比,我们还显示出显着的计算优势。
TRACE: Transport Alignment Conformal Prediction via Diffusion and Flow Matching Models
Department of Statistics and Actuarial Science University of Iowa(统计与精算科学系,爱荷华大学);Departments of Data Science and AI, and Applied Mathematics The Hong Kong Polytechnic University(数据科学与人工智能系、应用数学系,香港理工大学)
One-step Diffusion Models with Bregman Density Ratio Matching
LIX, École Polytechnique, CNRS, IPP(巴黎政治学院LIX研究所,法国国家科学研究中心,IPP);LIGM, École Nationale des Ponts et Chaussées, CNRS, IPP(巴黎综合理工学院LIGM研究所,法国国家科学研究中心,IPP)
Theoretical Guarantees for High Order Trajectory Refinement in Generative Flows
The University of Texas at Austin(德克萨斯大学奥斯汀分校);Stevens Institute of Technology(史蒂文斯理工学院);The University of Hong Kong(香港大学);University of Wisconsin-Madison(威斯康星大学麦迪逊分校);South China University of Technology(华南理工大学);The Simons Institute for the Theory of Computing(西蒙斯理论计算研究所)
DiffGED: Computing Graph Edit Distance via Diffusion-based Graph Matching
University of New South Wales(新南威尔士大学);University of Technology Sydney(悉尼科技大学);Zhejiang Gongshang University(浙江工商大学);Shanghai Jiaotong University(上海交通大学)
Straighter Flow Matching via a Diffusion-Based Coupling Prior
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络空间安全学院);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所)
Control Variate Score Matching for Diffusion Models
Google DeepMind(谷歌DeepMind);BIFOLD – Berlin Institute for the Foundations of Learning and Data(柏林学习与数据基础研究院);Machine Learning Group, Technische Universität Berlin(柏林技术大学机器学习组);Department of Artificial Intelligence, Korea University(韩国大学人工智能系);Max-Planck Institute for Informatics(马克斯·普朗克信息研究所)