평범한 필기장

Paper : https://arxiv.org/abs/2412.07517 FireFlow: Fast Inversion of Rectified Flow for Image Semantic EditingThough Rectified Flows (ReFlows) with distillation offers a promising way for fast sampling, its fast inversion transforms images back to structured noise for recovery and following editing remains unsolved. This paper introduces FireFlow, a simple yet effarxiv.orgAbstract Rectified Flow..

Paper : https://arxiv.org/abs/2209.03003 Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified FlowWe present rectified flow, a surprisingly simple approach to learning (neural) ordinary differential equation (ODE) models to transport between two empirically observed distributions π_0 and π_1, hence providing a unified solution to generative modelingarxiv.orgAbstract 본 논문..

Paper : https://arxiv.org/abs/2404.04650 InitNO: Boosting Text-to-Image Diffusion Models via Initial Noise OptimizationRecent strides in the development of diffusion models, exemplified by advancements such as Stable Diffusion, have underscored their remarkable prowess in generating visually compelling images. However, the imperative of achieving a seamless alignment betwearxiv.orgAbstract 모든 ra..

Paper : https://arxiv.org/abs/2411.16738 Classifier-Free Guidance inside the Attraction Basin May Cause MemorizationDiffusion models are prone to exactly reproduce images from the training data. This exact reproduction of the training data is concerning as it can lead to copyright infringement and/or leakage of privacy-sensitive information. In this paper, we present aarxiv.orgAbstract해결하려는 문제 D..

Paper : https://arxiv.org/abs/2407.21720 Detecting, Explaining, and Mitigating Memorization in Diffusion ModelsRecent breakthroughs in diffusion models have exhibited exceptional image-generation capabilities. However, studies show that some outputs are merely replications of training data. Such replications present potential legal challenges for model owners, espearxiv.orgAbstract문제 : 생성모델의 몇 o..